Researching People: The Collaborative Listener

advertisement
5
Researching People:
The Collaborative Listener
Portraying people depends on careful
listening and observing. In this chapter you
will learn to:
• use details to describe people
•
gather information from informants
• transcribe others' conversation
• borrow elements of fiction writing
Researching people means stepping in to the worldviews of others. When we
talk with people in the field or study the stuff of their lives-their stories, artifacts, and surroundings-we enter their perspectives by partly stepping out of
our own.
In an informal way, you are always gathering data about people's backgrounds and perspectives~their worldviews. "So where are you from?" "How
do you like it here?" "Did you know anyone when you first came here?" Not only
do you ask questions about people's backEthnography is interaction, collaboration.
grounds, but you also notice their artifacts
What it demands is not hypotheses,
and adornments-the things with which
which may unnaturally close study down,
they represent themselves: T-shirts, jewelry,
obscuring the integrity of the other, but
particular kinds of shoes or hairstyles. The
the ability to converse intimately.
speculations and questions we form about
-HENRY CLASSIE
others cause us to make hypotheses about the
people we meet. We may ask questions, or we may just listen. But unless we
listen closely, we'll never understand others from their perspectives. We need to
know what it's like for that person in this place.
Even in an informal conversation, we conduct a kind of ethnographic inter~
view. Good interviewing is collaboration between you and your informant, not
very different from a friendly talk. Listening to your relatives share stories at a
wedding, pouring over a photo album with your grandmother, and gossiping
about old times at a school reunion are all instances in which you can gather data
through collaborative listening. You've experienced your local media interviewing
219
220
Chapter 5
Researching People: The Collaborative Listener
sports heroes and newsworthy citizens, and you know famous television and
radio interviewers such as Oprah, Barbara Walters, and Charlie Rose. Your
fieldworking interviews might employ the same skills: establishing rapport, letting your informant digress from your questions, as well as carefully listening
and navigating the conversation process.
This chapter will help you strengthen the everyday skills of listening, questioning, and researching people who interest you. You'll experience interactive
ways to conduct interviews and oral histories. You'll look for and discover
meaning in your informants' everyday cultural artifacts. You'll gather, analyze,
write, and reflect on family stories. And you'll read some examples of how other
fieldworkers have researched and written about people's lives.
The Interview: Learning How to Ask
Fieldworkers listen to and record stories from the point of view of the informant- not their own. Letting people speak for themselves by telling about
their lives seems an easy enough principle to follow. But in fact, there are some
important strategies for both asking questions and listening to responses. Those
strategies are part of interviewing - learning to ask and learning to listen.
Interviewing involves an ironic contradiction: you must be both structured
and Hexible at the same time. While it's critical to prepare for an interview with a
list of planned questions to guide your talk, it is equally important to follow your
informant's lead. Sometimes the best interviews come from a comment, a story,
an artifact, or a phrase you couldn't have anticipated. The energy that drives a
good interview-for both you and your informant-comes from expecting the
unexpected.
Expecting the Unexpected It's happened to both of us as interviewers. As
part of a two-year project, Elizabeth conducted in-depth interviews with Anna,
a college student who was a dancer. Anna identified with the modern dancers at
the university and also was interested in anin1al rights, organic foods, and ecological causes. She wore a necklace that Elizabeth thought served as a spiritual
talisman or represented a political affiliation. When she asked Anna about it, she
learned that the necklace actually held the key to Anna's apartment-a much
less dramatic answer than Elizabeth anticipated. Anna claimed that she didn't
trust herself to keep her key anywhere but around her neck, and that information provided a clue to her temperament that Elizabeth wouldn't have known if
she hadn't asked and had persisted in her own speculations.
In a shorter project, Bonnie interviewed Ken, a school superintendent, over
a period of eight months. As Ken discussed his beliefs about education, Bonnie
connected his ideas with the writings of progressivist philosopher John Dewey.
At the time, she was reading educational philosophy herself and was greatly
influenced by Dewey's ideas. To her, Ken seemed to be a contemporary incarnation of Dewey. Eventually, toward the end of their interviews, Bonnie asked
The Interview: Learning How to Ask
md
our
letting
JeS-
tive
)Ver
yze,
ther
iforJout
ome
1ose
.ired
itha
your
tory,
a
res
~the
s. As
.nna,
:rs at
eco'itual
:, she
nu ch
lidn't
rmawnif
over
)nnie
~wey.
·eatly
ncar;sked
Ken which of Dewey's works had been the most important to him. "Dewey?" he
asked. "John Dewey? Never exactly got around to reading him."
No matter how hard we try to lay aside our assumptions when we interview others, we always carry them with us. Rather than ignore our hunches, we
need to form questions around them, follow them through, and see where they
will lead us. Asking Anna about her necklace, a personal artifact, led Elizabeth
to new understandings about Anna's self-concept and habits that later became
important in her analysis of Anna's literacy. Bonnie's admiration for Dewey had
little to do directly with Ken's educational philosophy, but her follow-up questions centered on the scholars who did shape Ken's theories. It is our job to
reveal our informant's perspectives and experiences rather than our own. And
so our questions must allow us to learn something new, something that our
informant knows and we don't. We must learn how to ask.
Asking
Asking involves collaborative listening. When we interview, we are not extracting information the way a dentist pulls a tooth, but we make meaning together
like two dancers, one leading and one following. Interview questions range
between closed and open .
Closed Questions Closed questions are like those we answer on application
forms or in magazines: How many years of schooling have you had? Do you
rent your apartment? Do you own a car? Do you have any distinguishing birthmarks? Do you use bar or liquid soap? Do you drink sweetened or unsweetened
tea, caffeinated or decaffeinated coffee? Some closed questions are essential for
gathering background data: Where did you grow up? How many siblings did
you have? What was your favorite subject in school? But these questions often
yield single phrases as answers and can shut down further talk. Closed questions can start an awkward volley of single questions and abbreviated answers.
To avoid asking too many closed questions, you'll need to prepare ahead of
time by doing informal research about your informants and the topics they represent. For example, if you are interviewing a woman in the air force, you may
want to read something about the history of women in aviation. You might also
consult an expert in the field or telephone government offices to request informational materials so that you avoid asking questions that you could answer for
yourself, like "How many years have women been allowed to fly planes in the
U.S. Air Force?" When you are able to do background research, your knowledge
of the topic and the informant's background will demonstrate your level of interest, put the informant at ease, and create a more comfortable interview situation.
Open Questions Open questions, by contrast, help elicit your informant's
perspective and allow for a more conversational exchange. Because there is no
single answer to open-ended questions, you will need to listen, respond, and
follow the informant's lead. Because there is no single answer, you can allow
221
222
Chapter 5
Researching People: The Collaborative Listener
yourself to engage in a lively, authentic response. In other words, simply being
interested will make you a good field interviewer. Here are some very general open
questions-sometimes called experiential and descriptive-that encourage the
informant to share experiences or to describe them from his or her own point
of view.
....................................................................................................................
Open Questions for Your Informant
Ii!
Tell me more about the time when ...
0
Describe the people who were most important to ...
&
Describe the first time you ...
0
Tell me about the person who taught you about. ..
0
What stands out for you when you remember ...
•
Tell me the story behind that interesting item you have.
@
Describe a typical day in your life.
®
How would you describe yourself to yourself?
0
How would you describe yourself to others?
....................................................................................................................
When thinking of questions to ask an informant, make your informant your
teacher. You want to learn about his or her expertise, knowledge, beliefs, and
worldview.
Box20
Using a Cultural Artifact in an Interview
iQt!l@•D
This exercise mirrors the process of conducting interviews over time with an informant.
It emphasizes working with the informant's perspective, making extensive and accurate
observations, speculating and theorizing, confirming and disconfirming ideas, writing up
notes, listening well, sharing ideas collaboratively, and reflecting on your data.
To introduce interviewing in our courses, we use an artifact exchange. This exercise
allows people to investigate the meaning of an object from another person's point
of view.
>ly being
'falopen
rage the
.vn point
eite1~1
choose a partner from among your colleagues. You will act as both interviewer and inforR
mant. Select an interesting artifact that your partner is either wearing or carrying: a key
chain, a piece of jewelry, an item of clothing. Both partners should be sure the artifact is
one the owner feels comfortable talking about. If, for example, the interviewer says, 11 Tell
..........
me about that pin you are wearing," but the informant knows that her watch has more
meaning or her bookbag holds a story, the interviewer should follow her lead. Once
you've each chosen an artifact, try the following process. Begin by writing observational
and personal notes as a form of background research before interviewing:
1.
Take observational notes. Take quiet time to inspect, describe, and take notes on
your informant's artifact. Pay attention to its form and speculate about its function.
Where do you think it comes from? What is it used for?
2.
Take personal notes. What does it remind you of? What do you already know about
things similar to it? How does it connect to your own experience? What are your
hunches about the artifact? In other words, what assumptions do you have about it?
(For example, you may be taking notes on someone's ring and find yourself speculating about how much it costs and whether the owner of the artifact is wealthy.) It is
important here to identify your assumptions and not mask them.
3.
Interview the informant. Ask questions and take notes on the story behind the
artifact. What people are involved in it? Why is it important to them? How does the
owner use it? Value it? What's the cultural background behind it? After recording
your informant's responses, read your observational notes to each other to verify or
clarify the information.
4.
Theorize. Think of a metaphor that describes the object. How does the artifact reflect
something you know about the informant? Could you find background material
about the artifact? Where would you look? How does the artifact relate to history or
culture? If, for example, your informant wears earrings made of spoons, you might
research spoon making, spoon collecting, or the introduction of the spoon in polite
society. Maybe this person had a famous cook in the family, played the spoons as a
folk instrument, or used these as baby spoons in childhood.
5.
Write. In several paragraphs about the observations, the interview, and your theories,
create a written account of the artifact and its relationship to your informant. Give a
draft to your informant for a response.
6.
Exchange. The informant writes a response to your written account, detailing what
was interesting and surprising. At this point, the informant can point out what you
didn't notice, say, or ask that might be important to a further understanding of the
artifact. You will want to exchange your responses again, explaining what you learned
from the first exchange.
7.
Reflect. Write about what you learned about yourself as an interviewer. What are your
strengths? Your weaknesses? What assumptions or preconceptions did you find that
you had that interfered with your interviewing skills? How might you change this?
8.
Change roles and repeat this process.
............
ant your
.efs, and
>p
•
223
Box20
continued
Lini Ge's Watch
Although we've done this exercise with many people-students, teachers, and
researchers-we're always surprised at the flood of cultural information that comes from
the pages people fill during their interviews. We both enjoy practicing interviewing with
artifacts, so we almost always participate in the process. Here's an interview Bonnie did
with her student Lini Ge:
When Lini handed me her watch, my first reaction was "Oh, she's a swimmer or some
other kind of an athlete." I noticed it was made out of a black, supple, strong, rubbery, plastic kind of material. It's very colorful and it says "sports watch" in yellow and
"water resistant" in purple on the strap.
There are three elliptical openings on either side of the strap, about three inches
on each side, suggesting, too, another kind of practical water resistance. "It's an ugly
belt," she tells me. "Do you call it a belt?" "Oh, strap," I answer; "it looks Ike a belt; it
has a black buckle, but I guess because you wear it on your wrist you call it a strap,"
and l realize it's yet another American synonym that would make no sense to a native
speaker of another language. It is, basically, a belt. Lini has owned it since her aunt in
Beijing gave it to her almost a decade ago, and she had to replace the strap when the
original one broke. She liked that strap better; it was black leather. I noticed that the
new strap has nine holes. When I had been a tourist in China, I learned the spiritual
and historical importance of the number 9. Lini laughed when l asked her about that
and said she'd never even bothered to count how many holes her watch strap had. I
realize it's the company's way to accommodate many sizes of wrists.
But truth to tell, such complicated watches scare me. l grew up on the old-fashioned
wind-up watch with a traditional face, and although I do love battery power, I still
prefer a round two-hand dial. My daughter changes all of my digital docks twice a year
because I find it so frustrating. But obviously, since Lini is a member of the high-tech
generation, all of the buttons and features would not be a problem for her. The mechanism and face are practical and complex, too, a slightly exaggerated square-about
1" x 1.3"-with many color-coded features, inside and outside of a thin red border.
She keeps the watch set to twenty-four-hour time, the convention used in China
(it is 18:20:54 when I observe it; 6:20 p.m. in American convention). But it's really not.
"I keep it about five minutes fast," she tells me. "I don't want to miss my bus, and I'm
always running behind." A critical skill for any student.
Lini has Lupus, I remember, a chronic illness. She also has problems with her
kidneys. Disease has changed the course of her life and studies. The more I observed
the watch before I talked with her, the more I realized that a major function for her
would be that she could keep track of her blood pressure-conveniently on her wrist
while also telling the time. Outside the border, there are four buttons, two on each
side. On the left, there is "adjust" and "mode," for keeping time. On the right, a black
button is labeled "bp start" and a yellow one labeled "restart." On the upper portion
of the face, aside the CASIO company logo, are the words "Blood Pressure Monitor"
and BP-120 (a model number, but ironically, the number of healthy blood pressure).
A yellow panel offers the day, the date, and a small red heart. The time is displayed
digitally: hour, minutes, and seconds. On the lower portion are the words "systolic"
224
and "diastolic," as well as what appears to be a sensor light on one side and a panel
that proclaims 11 pls" and "ekg." Aha. Medical terms about blood pressure.
I am intrigued that it's a Japanese watch with English language labels, owned by
a Chinese woman who got it as a gift from an aunt who doesn't speak English. It is
a symbol of our global economy, our multinational world. I assumed wrongly that
Lini must have bought it in the United States. "Oh, a Chinese person would buy this
watch this way," Lini told me. 11 Casio is a famous brand, and it's more convenient to
use these terms." Her friends in China were fascinated by the watch's special features, although Lini admits it isn't that convenient for its medical value. She quickly
discovered that the blood pressure features will only work with fresh batteries. But
her friends enjoyed taking their heartbeats and blood pressure with it. She hasn't yet
shared it with her U.S. friends. I want to buy her some fresh batteries and check to
see if it works for me!
I
I
Ii
~
I
In interviews, researchers sometimes use cultural artifacts to enter into the
informant's perspective. We might start by talking about something in our informant's environment: a framed snapshot, a CD or DVD collection, an interesting or unusual object in the roo1n-anything that will encourage co1nfortable
conversation. When we invite informants to tell stories about their artifacts,
we learn about the artifacts themselves (Lini Ge's watch) and, indirectly, about
other aspects of their world that they might not think to talk about. Artifacts,
like stories, can mediate between individuals and their cultures.
Learning How to Listen
Although most people think that the key to a good interview is asking a set of
good questions, vve and our students have found that the real key to interviewing
is being a good listener. Think about your favorite television or radio talk show
personalities. What do they do to make their informants comfortable and keep
conversation flowing? Think about someone you know whon1 you've always
considered a good listener. Why does that person make you feel that way?
Good listeners guide the direction of thoughts; they don't interrupt or move
conversation back to the1nselves. Good listeners use their body language to let
informants understand that their informants' words arc important to the1n, not
allowing their eyes to wander, not fiddling, not checking their watches or their
phones. They encourage response with verbal acknowledgments and follow-up
questions, with embellishments and examples.
To be a good listener as a field interviewer, you must also have structured
plans with focused questions. And you must be willing to change them as the
225
226
Researching People: The Collaborative Listener
Chapter 5
conversation moves in different directions. With open qt1estions, background
research, and genuine interest in your informant, you'll find yourself holding
a collaborative conversation from which you'll both learn. It is the process, not
the preplanned information, that n1akes an interview successful.
......................................................................................................, .............
Etiquette for Conducting an Interview
In addition to preparing yourself with guiding questions and good listening
habits, here are some basic rules of etiquette for conducting a successful inter~
view. Always keep in mind that you are using someone's time.
'*
Arrange for the interview at your informant's convenience. Your inter~
view should fit into that person's schedule, not vice versa. Put his or her
needs first.
®
Explain your project in plain language that your informant will understand.
Don't bore or scare them with insider expressions such as "ethnographic"
or 11 fieldsite research."
•
Agree on a quiet place to talk. Avoid places like cafes that have a lot of
ambient noise.
©
Arrive on time and be prepared. Make sure your equipment works (pens,
batteries, recording devices}. Have your questions and notepad ready.
©
Dress appropriately for the setting and for your informant. You'd wear
something different to interview a lifeguard on the beach than your grandmother in her living room.
®
Don't try to squeeze too much into a short time. Be sensitive to social cues
and, if necessary, arrange for an additional interview.
®
Thank your informant and follow up with a thank-you note, e-mail, or, if
appropriate, a token of gratitude.
....................................................................................................................
A Successful Interview
Paul Russ conducted illterviews with five AIDS survivors for an ethnographic
film, Healing without a Cure: Stories of People Living with AIDS, sponsored by a
local health agency. He developed a list of open and closed questions to prepare
for and guide his interviewing process. He knew that closed questions would
provide him with similar baseline data for all of his informants. For this reason,
he formulated some questions that had one specific answer:
Paul's Closed Questions
®
"How many months have you lived with your diagnosis?"
®
"When did you first request a 'buddy' from the health service?"
®
"Does your family know about your diagnosis?"
Learning How to Listen
nd
ng
10t
But the overall goal of his project was to capture how individuals coped with
their diagnoses daily, drawing on their own unique resources. He wanted to
avoid creating a stereotypical profile of a "day in the life of a person living with
AIDS" since he knew that no one AIDS patient's way of coping could represent
all other patients' coping styles. Paul constructed open questions to allow his
informants to speak from their lived experiences.
Paul's Open Questions
®
"What did you already know ahout AIDS when you were diagnosed?"
•
"How did others respond to you and your diagnosis?"
©
"What has helped you most on a day-to-day basis to live with the virus?"
©
"Have people treated you differently since you were diagnosed?"
In the following excerpt from his hundreds of pages of transcripts, Paul
talks with Jessie, a man who had been living with his diagnosis for eight years.
For Paul, this interview was a struggle because Jessie hadn't talked much with
others about AIDS. And because Paul chose to study people whose lives were
very fragile, he paid particular attention to the interactive process between
himself and his informants. In the following transcript, Paul uses Jessie's dog,
Princess, just as another interviewer might have used an artifact to get further
information:
P:
J:
iic
I
a
Lre
1ld
m,
P:
J:
What was your reaction when you were first diagnosed?
(This is one of the questions Paul posed to each of his five
informants. Because he was making a training film for public
health volunteers, he wanted to record people's initial reactions on
discovering that they had a publicly controversial illness.)
My first reaction? How am I going to tell my family. And I put it in
my mind that I would not tell anyone until it became noticeable.
And I wondered who would take care of me .... I knew sometimes
AIDS victims go blind. I panicked a little bit, and I started thinking
of all the things I have to do to make my life livable .... I started
thinking about the things I could do to make it go easier. And I
started thinking of things I would miss.
Like Princess, your dog?
(Paul knew froni previous talks that Jessie's dog was an important
part of his daily life.)
I've had Princess for three years. I had another red dachshund,
but she got away. I got Princess as a Christmas gift .... She
comforts me. She knows when I'm not feeling right. She comes
and rubs me. She goes places with me. If I'm in the garden,
she's right there. She can't let me out of her sight. Sometimes
227
228
Chapter 5
Researching People: The Collaborative Listener
I talk to her, late at night, we just lay there. She seems like
she understands .... I don't think she can live without me. If
something happens to me, she'll be so confused. I think she'll be
so lonely, she'll go off somewhere and just die .... I want to give
her to somebody. Maybe an older person, someone I believe will
take care of her.
(By talking about his dog, Jessie opened himself up to Paul. By
folloiving up on Jessie's comment about "things he'd nziss," Paul
deepened their interaction and intensified their talk. It was not the
dog herself that was important in this exchange but what Princess
represented fro1n Jessie's perspective. Paul did not intend to make
Jessie talk about his fear of dying, but it happened naturally as
he talked about Princess. At this point, Paul found a way to ask
another one of the prepared questions that he used with each
of his informants. And Jessie's answer brought them back to
Princess.)
P:
What's your typical day like?
J:
My typical day is feeding Princess, letting her out, doing my
housework. I like to do my work before noon because I'm addicted
to soap operas .... I like to work in the yard. I've got a garden. I
have some herbs. And I like every now and then to pray. I go to the
library. I do a great deal of reading.
(Paul continued to interview Jessie about his spirituality and his
reading habits. He brought this interview around to another preplanned question that he asked of all his AIDS infonnants.)
P:
What advice do you have for the newly diagnosed?
J:
Don't panic. You do have a tendency to blow it out of proportion.
And find a friend, a real friend, to help you filter out the negative.
Ask your doctor questions. Let it out and forgive. Forgive yourself,
you're only human. And forgive the person you think gave it to you.
Then you will learn that the key to spirituality is to abandon
yourself .... I don't want a sad funeral. I want music, more music
than anything e!Se. I don't want my family to go under because of
this disease.
Paul's interviews eventually became a training fihn for volunteers al the
Triad Health Project and area schools that wanted to participate in AIDS support and education. In the film, Paul has the advantage of presenting his data,
not just through verbal display but visually as well. As Paul conducts his interviews, we hear his voice and see his informants-their surroundings and artifacts, their gestures and body language, and the tones of voices as they respond
to Paul.
Box21
Establishing Rapport
Paul Russ worked hard to establish rapport with his informant. Rapport doesn't happen
in one short interview. Interviewing is a collaborative and interactive process in which
researchers make themselves knowledgeable about their informants' positions, interests,
feelings, and worldviews.
In this activity, you will reflect on your relationship with an informant and gain greater
understanding of yourself as a researcher. Write a short paper about your subjective
attitude toward an informant. Think about whether you've felt tentative or hesitant
toward your informant, feelings that you may not want to write about in your final paper
but that you acknowledge and understand as part of your researcher self. Use the followM
ing list to guide you:
1.
Describe your first meeting with your informant. What did you notice about yourself
as you began the interview process?
2.
Describe any gender, class, race, or age differences that may have affected the way
you approached your informant.
3.
Discuss ways you tried either to acknowledge or to erase these differences and the
extent to which you were successful.
4.
Discuss how your rapport changed over time in talking with and understanding your
informant and her worldview.
1113!!0i"®'~'111
Paul Russ faced many race and class differences when he interviewed his informants
about how they lived with AIDS. The most obvious was health, since his informants were
facing disease and he was not. Paul's response describes the many conflicting feelings he
had when he interviewed Jessie:
I picked up Jessie to drive him to the Health Project office for the interview. At first,
we didn't conduct the interviews at his house. I'm not sure if he was uncomfortable
about me seeing the inside of his house, if he didn't want the neighbors seeing a tall
white guy carrying a bunch of camera equipment into his house. Anyway, as Jessie
rode in my car, I was incredibly aware of the two different worlds we came from. I
had a bad case of white man's guilt. As he sat in my car, I apologized for the dog hair
left from taking my two dogs to the vet. He said that it was fine, that he was used to it.
Then he mentioned his dog, Princess. It was the first thing we had to talk about. Jessie admitted that he had little family support to cope with AIDS and that Princess was
his family. I shared that my dog had had a difficult pregnancy and that I almost lost
her. That's when he first opened up to me about his fear of living without Princess or
Princess living without him. When it later came up in our interview, it was an obvious
opportunity to encourage Jessie to speak personally.
229
Box21
continued
It was essential to establish common ground with him because I felt I had nothing in common with Jessie. Perhaps this was because he did not come from where
I came from and, perhaps, because he did not look like me. And while I've never
considered myself prejudiced, I realize that we all have prejudices deeply buried
inside no matter how intelligent or informed we are. In order to know him with some
degree of intimacy, I had to be vulnerable and share myself. I had to address the
baggage of race, class, education. I did this with all the informants in my project, and
it scared me because being friends with someone who is facing mortality requires an
emotional investment. I knew I had to establish a friendship.
While I was making a personal connection with Jessie, I also had professional
distance. With everything that came out of Jessie's mouth, I was thinking about how
it could be used in the final project. For me, interviewing is very active. It's not passive at all. You have to listen for meaning and listen for what's not being said. I had
trouble getting Jessie to speak from the heart. His responses to early questions were
pressed. I knew that if I were writing his story for a reader, l could project a much
clearer sense of his identity than he gave me on camera. I knew that. But I wasn't
writing his story. My mission was to record him telling his story in his own words.
So I looked for opportunities to help him reveal himself to me. Princess was one of
these opportunities.
Recording and Transcribing
Interviews provide the bones of any fieldwork project. You need your informants'
actual words to support your findings. Without informants' voices, you have no
perspective to share except your own. When you record and transcribe your
interviews, you bring to life the language of the people whose culture you study.
The process of recording and transcribing interviews has been advanced by
computers, software, and audio recorders that are small, relatively inexpensive,
and easily available. It's no coincidence that interviewing and collecting oral histories have become more :Popular in recent years \Vi th these accessible technologies. With a counting feature to keep track of slices of conversation and a pause
button to slow down the transcription process, even the most basic recorder
becomes a valuable tool for the interviewer.
Your choice of recording device will partially depend on whether you need
a recorder that is lightweight and handheld, high resolution, and low noise, or
one that records to a memory card. You may find that an inexpensive recorder
will suit your purposes just fine if you want to record one-on-one in a quiet space
and transcribe it yourself before integrating the interview into your own text.
However, if you plan to record in a noisy setting and make both the audio and
written transcription available, you will need a higher-quality recording device.
230
Recording and Transcribing
How you are going to share your work is another consideration. Will you
transcribe it into print only, or will you also distribute it as an audio file, podcast, or part of a multimedia presentation? How you will use your interview
material will affect what kind of equipment you purchase or rent.
Advances in recording devices and software have cut down the tedium of
sorting, classifying, and organizing huge amounts of data. But transcribing is
tedious business nonetheless. Three or more hours might be required to transcribe one hour of recording. And editing your audio files (if you choose to do
that) will require even more time. However, you learn an enormous amount
about your informants and yourself as you listen, replay, and select the sections
for your study.
You don't want to record everything you hear, nor do you want to transcribe it all. That's why it's important to prepare ahead-with research, guiding questions, and adequate equipment as well as knowledge of how to use it.
The following guidelines will help make your recording and transcribing go
smoothly.
1!"
Prepare your equipment Dead batteries and full memory cards can
ruin your data collection. Always can)' extras. Test your recorder before
using it by stating your name, the date, the place of the interview, and the
full name of your informant, and then playing this information back to
yourself. If you use a microphone, check out its range before you begin.
Most fieldworkers have stories of losing interviews because of equipment
malfunction. Be prepared.
)
r
y
"
Plan to take notes Consider how you will take fieldnotes during the
interview so that you'll capture all the features of the experience and have
a backup in case your recording equipment fails. You want to note the
environment where the interview takes place, the facial and body language
of your informant's responses, and any hesitations or interruptions that
take place. Your fieldnotes will help supplement the actual recording. Also
consider taking photos that you can use later to jog your memory.
;;~
Organize your i11terview time Be considerate in setting up a time and
place for the interview. Ask your informant what's convenient for him or
her. Arrive a few minutes early and test your equipment as well as the space
so that you don't have any extraneous noise or distractions. Remember to
have a timepiece-be it a watch or a phone-so that you can keep track of
interview time.
e
r
:I
r
r
e
t.
:I
Obtain your equipment Before borrowing or purchasing quality
equipment, research what's available. A digital counter that helps keep
track of time, multidirectional microphones that minimize ambient noise,
and functions that record separate tracks are some key features that will
facilitate your research. Investigate what's available and appropriate to
your research.
231
232
Chapter 5
Researching People: The Collaborative Listener
Organize time to listen to your audio
Begin the labeling process as
soon as possible. Key the filename (date, place, and informant's name)
into your audio device or download the file from your memory card
and label it on your computer. Also write down the filename of each
recording in your fieldnotes. After the intervie\.v, listen to the recording
as soon as you can to keep it fresh in your inind. Take notes as you
listen for topics covered, the1nes that emerge, and possible follow-up
questions.
Transcribe the interview As soon as you can, begin to transcribe
your recording. Don't wait too long, as the initial listening process
enhances your memory of the interview and your sense of purpose
about the project. Remember that you do not need to transcribe all the
material you record-only the sections that arc useful to your study.
To transcribe, listen directly fron1 the recorder or upload your audio
files to your computer. Check with your media lab to see if they have a
device sometin1es known as a footswitch. If they do, you can attach it to
your audio device to stop and start the recording as needed. This can be
immensely helpful. Whatever sections you decide to use, transcribe them
word for word using parentheses or brackets to indicate pauses, laughing,
interruptions, sections you want to leave out, or unintelligible words. For
example: [Regina laughs nervously] or [phone rang, maybe match? Or
march?-check with Regina] or [unintelligible word].
Bring yoi1r infonnant's la11guage to life As a transcriber, you must
bring your informant's speech to life as accurately and appropriately as
you can. Most researchers agree that a person's grammar should remain
as spoken. If an informant says, "I done," for example, it's not appropriate
to alter it to "I did." But if when you share the transcript with your
informant she chooses to alter it, respect those changes. As well,
many characteristics of oral language have no equivalents in print.
For more on using insider
language in your writing,
It is too difficult for either transcriber or reader to attempt to
see Chapter 6, page 2a-1.
capture dialect in written form. "Pahk the cah in Hahvahd Yahd"
is a respelling of a Boston accent, meant to show how it sounds.
But to a reader who's never heard it~even to an insider Bostonian who
isn't conscious of her acc~nt-the written version of her oral dialect
looks artificial and complicates the reading process. Anthropologists and
folklorists have long debated how to record oral language and currently
discourage the use of spelling as a \vay to approxhnate oral language.
Share your transcript Offer the transcripts to your infor1nant to read
for accuracy, but realize that you won't get many takers. Most informants
would rather wait for your finished, edited version of the interview. In any
case, the informant needs the opportunity to read what you've written. In
some instances, the informant may make corrections or ask for deletions.
But most of the time, the written interview becomes a kind of gift in
exchange for the time spent interviewing.
Recording and Transcribing
Edit the audio files
You may decide to go beyond the written transcript
to include your raw audio recording in a Web~based media presentation
that will add an extra dimension to your project. To create a usable
audio clip, you will need to use software to edit or splice your recordings
together. Talk to fellow students or researchers about what they use. If
possible, test-drive the editing software in order to understand its full
capacities before you purchase it.
You'll also want to consider what medium in which to share your audio
clips, as this will affect their length and content. Biogs, multimedia presentations, and podcasts are some of the popular formats at the time of this writing.
Be sure to include the URL for your blog, Web site, or podcast in your written
work-especially if you plan to publish it-so that your readers can access it
and enhance their understanding of your study.
:· .............................. "
.................. ., '.' ....................................................... ' ...:
Reminders for Recording and Transcribing
Set up your interview.
Familiarize yourself with your recording device.
:0
Arrive early and evaluate your recording environment.
Conduct the interview and take notes.
.
"P
Arrange sufficient time for listening to and transcribing the interview.
('!
Consider your final presentation format and its audience.
. . . . . . . . . . . . . . . 00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,
............................................ .
.
Fieldworkers must turn interview transcripts into writing, n1aking a kind
of verbal film. As interesting as interview transcripts are to the researcher,
they are only partial representations of the actual interview process. Folklorist
Elliott Oring observes, "Lives are not transcriptions of events. They are artful
and enduring symbolic constitutions which demand our engagement and
identification. They are to be perceived and understood as wholes" (258). To
bring an informant's life to the page, you must use a transcript within your own
text, son1etimes describing the setting, the informant's physical appearances,
particular mannerisms, and language patterns and intonations. The transcript
by itself has little meaning until you bring it to life.
Cindie Marshall conducted a semeste1,long field project at Ralph's Sports
Bar, freq11ented by 1nen and \vomen who ride motorcycles and describe themselves as bikers. In "Ralph's Sports Bar," she combines her skills as a listener,
an interviewer, and n1ost of all a writer. In her study, her informants speak in
their O\vn voices, but Cindie contextualizes them, offering readers a look into
the biker subculture as it exists at Ralph's. As you read Cindie's research study,
notice the fieldworking skills she brings together.
233
234
Chapter 5
Researching People: The Collaborative Listener
Ralph's Sports Bar
Cindie Marshall
The Arrival
Everyone in the bar eyed me suspiciously during my visit. I didn't dress, walk,
or talk like the other women. Because of the illegal activity that goes on in the
bar, I guessed they were probably wondering if l was an undercover cop. I
realized that given the way the groups marked their boundaries, they probably
wouldn't be very inviting to a stranger asking a lot of questions about their
11
culture."
I decided that I would go to the table out front and perhaps there I could
find someone to answer my questions. At the very least, I could get some
fresh air and think about my approach. When I arrived at the table, there was
a woman sitting alone, drinking a beer. She had obviously had a lot to drink.
But we talked for a minute, and she was really quite nice. I introduced myself,
and she said her name was Teardrop. It was dark outside, so I couldn't see her
very well. We talked about the weather, and that led her to tell me that she had
moved here from Michigan three years ago and that she came to Ralph's
every day.
Ralph's Sports Bar
Recording and Transcribing
I knew that this was my opportunity to talk to a patron. On a long shot, I asked
her if she would like to shoot a game of pool. She agreed, and I was delighted. This
was an opportunity to ask questions and be seen with a regular.
Once we were inside, I saw that Teardrop was wearing new Lee jeans, a nice
pale yellow sweater, and a heart-shaped brooch. Her hair was brown but was
showing signs of graying. She had it neatly pulled back, and her bangs were teased
and carefully sprayed in place. It wasn't until she laughed that I noticed she was
missing her front teeth, both top and bottom. She had a small dark-green vine tattoo around her wrist. Most of the tattoo was covered by her watch.
As we played pool, I noticed she also had a black tattoo in the shape of a teardrop high on her cheek. When I asked her about her teardrop tattoo, she started
really talking to me. She told me that when she was 13, she had been kidnapped
by a group of bikers. The biker that kidnapped her had eventually sold her to a
fellow biker. This went on for years until finally, three years ago, she got away. The
teardrop was there because she couldn't cry anymore.
After hearing Teardrop's story, my admiration for the bikers' "do your own
thing" attitude was lost. She had described sheer abuse, and she wore that abuse
both on her face, in the shape of a teardrop, and in her smile, which was darkened
by missing teeth.
I left the bar to reflect on all that I had seen. I wanted to know why the groups
in the bar would come together in a place just to keep themselves segregated. The
only way I could get the answers to my questions was to talk to a person who had
been in all three groups and had spent a lot of time at the bar: Alice.
Conversation with Alice
Alice's boyfriend, Ralph, owned the bar. She had worked there prior to becoming
the receptionist at the law firm at which I worked, and she could probably tell me
everything I wanted to know.
Alice agreed to my interview, and I prepared a list of the three groups,
outlining what I thought their characteristics were. Alice read my list and we
began.
Characteristics of the "Rednecks"
We both agreed that one of the groups we would call "rednecks" for lack of a
better term. My list ascribed the following characteristics to this group: they
would be lazy; they would value freedom; they would not like or adhere to any
rules imposed on them; they would have no self-pride, either in their work or
appearance; they would demean women; they vvould have no materialistic values;
they would have no work ethic; and they would have no moral code among
themselves-it would be every man for himself.
After reviewing my list, Alice commented that actually, "they are hardworkR
ing and take pride in their jobs. Because they like to be able to say 'I do something
well.' ... Most of them are blueRcollar workers-construction workers, electricians,
235
236
Chapter 5
Researching People: The Collaborative Listener
Ralph's Sports Bar parking lot
people who do things with their hands. They're good at what they do to a certain
extent. A lot of them do change jobs frequently because of the drinking problem
that they have, and I think the majority of them do have drinking problems ....
They are lazy in the sense that they don't aspire to be anything more than what
they are."
We discussed how they treated their women, and Alice was quick to point out
"that's the biggest thing that they do .... It makes them feel like they are bigger."
We had talked prior to the interview about how uncomfortable I was at this bar. It
seemed that all the men treated the women with little or no respect.
On the subject of the rednecks' commitment to their fellow rednecks, I felt
that it would be every man for himself. I doubted that the rednecks had lasting
bonds or would stand up for a buddy in trouble. Alice said that she had "seen
where one night two guys would sit there and be buddy-buddy and would fight
together side by side. Two weeks later, they would fight each other."
Overall, the only real corrections Alice made to my list of characteristics was
that she strongly disagreed with my idea that rednecks were lazy. I asked her later
outside the interview how they could all be working yet spending their entire day
at Ralph's. She told me those individuals worked third shift and would simply go to
work drunk.
Characteristics of the Bikers
I had made a similar list of characteristics for the bikers at Ralph's. I had decided
that they, too, would value freedom and not like rules imposed on them. However,
they would have a higher moral code among themselves. I guessed that they didn't
care what other people thought of them and therefore would not be materialistic.
They would likely believe in making their own way and have a higher work ethic
than the rednecks. I would categorize them as more the "weekend warrior" type. I
also concluded that unlike the rednecks, they were probably ritualistic in the way
T
I
Recording and Transcribing
they meticulously parked their bikes. As for their treatment of women, after talking
with Teardrop, my view on that was clear. To the bikers, women were property, and
that was all.
Alice started her revision of my list by first telling me that 11 th ere are even
two different classes of bikers .... There are some bikers that are construction
workers, moochers, lowlifes. They live from day to day in how they get their
money, how they live. They come from a culture of brotherhood and "my buddy."
They are more clannish in that, if a guy drives a bike, they will stand beside
him no matter what. But also, there are a lot of bikers that come in there that
are white~collar workers; they do this on the weekend and like to be someone
different. They change their persona and how people perceive them. They put
on their leather pants and leather jackets and ride these motorcycles all weekend long and go back to their jobs. There are four or five of them that nobody
knows .... They are businessmen who own companies over in High Point ... but
you would never guess it by the way they look when they are sitting there on the
weekends drinking."
When we talked about the stereotypical biker, I asked her if she felt they
"treat women the same way the rednecks do." She said no. It was more of a kind
of ownership. She said, "'My old lady' and 'my old man,' this is the way they talk.
Rednecks are a little more respectful .. .'this is my wife' or 'this is my girlfriend: I
think that the 1old lady' and 1old man' can change from week to week. I have seen
[bikers] swap women around as if they were pieces of property." Surprisingly
enough, she made her point about the way bikers treat women by telling me
Teardrop's story.
We talked about the bikers and how they felt about freedom. Alice pointed
out, "You will find that most of the bikers will be outside, they will not be inside
the bar.... Bikers like the openness and the freedom -that's why they're bikers."
Conclusion
Alice is a reliable source of information on all three groups because she has been
very deeply connected to all of them. She truly does cross the line of the cultural
boundaries. As she put it, 11 l1ve been there. I've been a drinker and down on my
luck and slept in my car. I've been just where they are on many occasions ...."
My own stereotypes show in my list.of characteristics that I mapped out for
Alice to review. As for Alice, it is clear that she has stereotyped these groups, too.
After talking with her and visiting the bar, my question still remained: Why
would these distinctly different groups of people, each representing a unique
culture, come to one small bar, each mapping their turf and intentionally stayw
ing separated? Alice felt that the common bond was Ralph. Ralph moved easily
between the cultures and was actually a part of each group. She said that he did it
much the same as she did, the difference being that Ralph held the role of "leader"
or "policer" of each group. It was because of Ralph that the groups could all drink
their beer in harmony while at the bar.
I feel that it goes much deeper than that. The one common bond that all of
these groups share is love of freedom. The whitewcollar, partwtime biker enjoys the
237
238
Chapter 5
Researching People: The Collaborative Listener
freedom of wearing his leather and riding his Harley on the weekend without
anyone knowing. The redneck enjoys the freedom to drink his beer and be totally
wild if he wishes. All these people go to Ralph's to escape, to be free of
Read Cindie Marshall's full
the
watchful eyes of a judgmental society. They like the comfort of not
essay at bedfordstmartins
.com/fieldworking, under
worrying what anyone else will think because they know no one will
Student Essays.
care ....
Commentary: With More Time , , ,
I would love to have had more time to visit this bar and try to become part of
this culture. I am certain that there are errors in my list of characteristics, and I
know that these characteristics do not apply to every individual. It would have
been interesting to make connections in the bar and test my ideas; to prove
or disprove the stereotypes that I have about each group. With that research
perhaps it would be more clear where the stereotypes actually come from. I have
asked myself that question over and over again while writing this paper. At this
point, I cannot clearly say where they come from. I think it is a combination of
many sources.
I am still very intrigued with the biker culture. I wish I could have talked with
more female bikers to get their ideas on the treatment-of-women issue. Teardrop
really changed my attitude.
• • • • • o 0 0 0 0 0 0 0 0 0 0 0 0 0 ' 1 0 > 0 0 0 0 0 0 0 • • • ; o o o o 0 0 0 0 0 0 0 0 0 0 , , q o o 0 0 0 0 0 0 0000;00•<>0000~000~••0000000000•00000
Before we discuss Cindie Marshall's study, we want to show you some
excerpts from her portfolio that illustrate her notetaking and categorizing skills.
In some of her interviews, she simply took notes and later checked them with
Alice. But she also recorded other conversations with Alice.
As she learned more from her interviews with Alice, Cindie looked carefully
at the words in the interview and began to see the patterns in the biker culture
that Alice described. Here is a sample from her recorded interview.
Cindie Marshall's Interview Notes
INTERVIEW WITH ALICE
C: Let's go over the category list. You believe the rednecks value their freedom
and are hardworking.
A: I think that they are hardworking and take pride in their jobs. Because they
like to be able to say, "!do something well." But their whole goal in what I
have viewed is that "I work to drink and I drink to work."
C: What kind of work do they do normally?
A: Most of them are blue-collar workers-construction workers, electricians,
people who do things with their hands. They're good at what they do to
a certain extent. A lot of them do change jobs frequently because of the
Recording and Transcribing
ly
I
drinking problem that they have, and I think the majority of them do have
drinking problems.
C: We talked about their being lazy, and we concluded that they're lazy in the
sense that they don't aspire to be anything more than what they are.
A: Yes. If their daddy taught them how to do construction or to build houses,
they don't make goals other than that.
C: And no self-pride in terms of their appearance?
A: Yes.
C: And we talked about how they demean women.
A: Yes, very much so. That's the biggest thing that they do. It makes them feel
stronger and feel like better people. It makes them feel like they are bigger.
ve
}
me
C: We talked about every man for himself, and you were saying that there were
no lasting bonds between the members of this particular culture within
the bar.
A: This is true. Because I have seen where one night two guys would sit there
and be buddy-buddy and would fight together side by side. Two weeks later,
they would fight each other. If the situation changes. The alcohol changes
people's personalities, and it varies day to day who is buddy-buddy.
C: The other culture that I saw when I was there is-you have your rednecks
and you have your bikers. What is the difference between the bikers and the
rednecks?
lls.
ith
A: I think that there are even two different classes of bikers.
illy
Jre
A: In this bar. In the sense that there are some bikers that are construction
workers, moochers, lowlifes. They live from day to day in how they get
their money, how they live. They come from a culture of brotherhood and
"my buddy." They are more clannish in that, if a guy drives a bike, they will
stand beside him no matter what. But also, there are a lot of bikers that
come in there that are white-collar workers; they do this on the weekend
and like to be someone different. They change their persona and how people
perceive them. They put on their leather pants and leather jackets and ride
these motorcycles all weekend long and go back to their jobs. There are four
or five of them that nobody knows. They are very honest and they tell you
what they do, that they are businessmen who own companies over in High
Point and one thing and another. But you would never guess it by the way
they look when they are sitting there on the weekends drinking. (Need to
ask her how these guys are treated by the full-time bikers. Do they tell the
full-time bikers about their other identity?)
m
ey
I
C: In this bar?
C: So those particular guys really fit both in among the bikers and the whitecollar professionals.
239
240
Chapter 5
Researching People: The Collaborative Listener
A: Yes.
C: So they are just choosing what culture they want to be in on weekends
basically. They could go into the biker thing or the professional white-collar
worker thing if they wanted to, just whichever way they wanted to go.
A: Yeah.
One of the strengths of Cindie's study is that she acquires different perspectives about Ralph's Sports Bar, which mirror the three categories of patrons she
is attempting to understand. Cindie positions herself at the outset as part of the
"white-collar weekend professional" group, admitting her own stereotypes. The
second perspective is what Cindie calls the "regular bikers," and her informant
Teardrop has belonged to that "regular hiker" culture much of her life. The third
perspective comes from Alice, Cindie's colleague at the law firm and Ralph's girlfriend, who defends the "redneck" position.
Cindie was fortunate to find Teardrop, who was in many ways an ideal informant. Teardrop frequents Ralph's every day, so she is an insider to the culture.
But Teardrop had stepped out of the biker culture long enough to be able to
reflect on it Teardrop was also an ideal informant because Cindie interacted
with her informally, gathering data as they shot pool together. Cindie didn't
record Teardrop's story but remembered it and wrote it into her ficldnotes. She
didn't worry about forgetting Teardrop's story because it was indelible, just like
the tattoo that was there "because she couldn't cry anymore." Cindie learned
this story from Teardrop as they interacted, each gaining trust for the other. Had
Cindie pulled out her tape recorder and tried to interview Teardrop by asking,
"So tell me how you got your name?" she probably wouldn't have heard Teardrop's story. The process of interaction and rapport allowed Cindie to acquire
good insider data.
Cindie's interview with Alice was entirely different because it was stn1ctured and planned. Cindie prepared a list of questions based on her own fieldnote observations of the three categories of patrons at the bar. Because Alice
was already a friend from the law firm where Cindie worked, she didn't need to
establish rapport.
Both Teardrop's and Alice's perspectives give weight and evidence to Cindie's own field observations, allowing her to confirm and disconfirm her data. As
we point out throughout this book, when researchers use multiple data sources
(interviews, fieldnotes, artifacts, and library or archival documents), they triangulate their findings. In this semester-long study, Cindie collected and analyzed varied sources (interviews, fieldnotes, and artifacts), and wrote from the
perspective of a won1an who intervievved other women about a predominantly
male subculture.
Because Cindie's final account is smoothly written and her research skills
are well integrated, it might not be apparent that she has engaged in many
Recording and Transcribing
aspects of the fieldworking process. A summary of her many fieldworking strate·
gies reveals, however, that Cindie was able to do all of the following:
r
Prepare for the field.
She gained access through her colleague Alice, an insider at Ralph's.
She read other research studies and material about fieldworking.
She drafted a research proposal that explained her interest in the biker
C·
subculture.
1e
1e
1e
She wrote about her assumptions and uncovered her prejudices about
bikers and rednecks.
:>
0
nt
:d
·1.
Use the researcher's tools.
She established rapport by hanging out at Ralph's and by locating
Teardrop, an insider informant.
She observed, taking detailed fieldnotes about the physical environment.
She participated in the culture by shooting pool with Teardrop, talking and
interacting at the same time.
She gathered descriptions of many cultural artifacts, taking photos and
noting what they implied.
She interviewed two informants, Alice (formally) and Teardrop
(informally), taking fieldnotes on their physical characteristics as well as
their stories.
g,
.rre
C·
:1.
:e
to
rr-
\s
'5
·i-
a·
1e
She transcribed Alice's interview, selecting potential sections to use in her
final project.
Interpret the fieldwork .
She read her data, proposal, fieldnotes, interview notes, and transcripts,
looking for themes and patterns.
She categorized her data into findings according to the three groups she
observed.
She made meaning co1laboratively with her informants:
With Alice, she verified her categories and disconfirmed her own
cultural stereotypes in her interview.
With Teardrop, she reflected on her data, particularly after her
interview, and expanded her findings with insights about the treatment
of women within the biking culture.
She reflected on how her data described a subculture.
ly
Present the findings.
She acknowledged and wrote about her position as an outsider.
ls
1y
She used descriptive details in her writing, selecting particular written
artifacts that convey the meaning of the culture.
241
Researching People: The Collaborative Listener
Chapter 5
242
She turned her informants into characters by melding her fieldnotes and
her transcripts.
She integrated her informants' voices with her narrative by using direct
quotations from her transcripts.
She designed sections with subheadings to guide her reader through
the project: "The Arrival," "Conversation with Alice," "Characteristics of
the 'Rednecks'," "Characteristics of the Bikers," "Conclusion," and her
commentary called "With More Time .... "
Analyzing Your Interviewing Skills
Reviewing and analyzing an excerpt from your transcripts can help you refine your interviewing skills and see ways to improve them. Pausing to look closely at your interviewing
and transcribing techniques may smooth the way for the rest of your project.
Select and transcribe a short section (no more than a page) from a key interview to share.
Play the corresponding portion of the recording as your colleagues read the transcript
and listen. Have your colleagues jot down notes and suggestions about your interview so
that you can discuss their observations together afterward:
1P
Has the interviewer established rapport with the informant?
0
Who talks more, the informant or the interviewer? Does that seem to work?
Bi
What was the best question the interviewer asked? Why?
©
What question might have extended to another question? Why?
e
How did the interviewer encourage the informant to be specific?
@
Were any of the questions closed?
Try using the line-by-line analysis in a small section, like the following example from
Cindie Marshall.
Here we analyze one of Cindie's transcripts, a portion of her interview with Alice. In
this excerpt, both interviewer and informant struggle with what they mean by the word
redneck and its associated cultural stereotypes. Fieldworkers need to be sensitive
to words that have different meanings for insiders in a culture. From her outsider
perspective, Cindie knew that the word redneck is a loaded term and that it's used
differently in different areas of the country. She was eager for Alice to help her clarify
what it meant to different groups at Ralph's Sports Bar.
C:
You believe the rednecks value their freedom and are hardworking.
(Cindie wants confirmation from Alice about one group of people she has observed
A:
frequenting Ralph's Sports Bar. She tries to get Alice to untangle her own perspective about
this category of patrons in the bar.)
I think that they are hardworking and take pride in their jobs. Because they like to be
able to say, "I do something well." But their whole goal in what I have viewed is that"!
work to drink and I drink to work."
C:
What kind of work do they do normally?
(Cindie follows Alice's lead, asking for more information, trying to find out more about
what they each mean when they use the term redneck. When the researcher recognizes
that one word has different meanings among different informants, she ought to try to
understand it.)
A:
Most of them are blue-collar workers-construction workers, electricians, people
who do things with their hands. They're good at what they do to a certain extent. A
lot of them do change jobs frequently because of the drinking problems that they
have, and I think the majority of them do have drinking problems.
C:
We talked about their being lazy, and we concluded that they're lazy in the sense that
they don't aspire to be anything more than what they are.
(Rather than trying to get more information from Alice about biker patrons who "do things
with their hands," Cindie introduces her own stereotype-that bikers are lazy, an idea that
she and Alice had discussed before. Cindie's question represents her indecision about
whether her stereotypes of bikers were true or if they came from movies.)
A:
Yes. If their daddy taught them how to do construction or to build houses, they don't
make goals other than that.
(Alice tried to move the conversation away from yet another stereotype-that bikers are
lazy-by offering her own observation: that bikers see1n to have limited career goals. Both
interviewer and informant struggle to understand each other's stereotypes.)
C:
And no self-pride in terms of their appearance?
(This is a leading question and in the end a closed question. Alice has no choice but to
answer yes or no. Later in the itlterview, Cindie asks a descriptive question that prompts
Alice to talk about what bikers actually do like to wear.)
A:
Yes.
C:
And we talked about how they demean women.
(Cindie raises yet another topic based on earlier conversations with Alice and also
based on stereotypes of bikers that, in her later interview with Teardrop, prove to be
true.)
A:
Yes, very much so. That's the biggest thing that they do. It makes them feel stronger
and feel like better people. It makes them feel like they are bigger.
243
246
Chapter 5
Researching People: The Collaborative Listener
them and talking to them, while they were killed. She was very distressed at their
deaths-"1 wept and wept:'
She had just finished the story when we arrived at her home-a small twostory town house, some distance from the campus. Downstairs was comfortable,
with the usual amenities-a sofa, armchairs, a television, pictures on the wall-
but I had the sense that it was rarely used. There was an immense sepia print of
her grandfather's farm in Grandin, North Dakota, in 1880; her other grandfather,
she told me, had invented the automatic pilot for planes. These two were the
progenitors, she feels, of her
agricultural and engineering
talents. Upstairs was her study,
with her typewriter (but no word
processor), absolutely bursting
with manuscripts and booksbooks everywhere, spilling out
of the study into every room
in the house. (My own little
11
house was once described as a
machine for working," and I had
a somewhat similar impression
ofTemple's.) On one wall was
a large cowhide with a huge
collection of identity badges
and caps, from the hundreds of
conferences she has lectured
at. I was amused to see, side by
Temple Grandin, Ph.D., a gifted animal scientist
side, an l.D. from the American
Meat Institute and one from the
American Psychiatric Association. Temple has published more than a hundred
papers, divided between those on animal behavior and facilities management
and those on autism. The intimate blending of the two was epitomized by the
medley of badges side by side.
Finally, without diffidence or embarrassment (emotions unknown to her), Temple showed me her bedroom, a:·n austere room with whitewashed walls and a single
bed and, next to the bed, a very large, strange-looking object. "What is that?" I asked.
"That's my squeeze machine," Temple replied. "Some people call it my hug
machine."
The device had two heavy, slanting wooden sides, perhaps four by three
feet each, pleasantly upholstered with a thick, soft padding. They were joined by
hinges to a long, narrow bottom board to create a V-shaped, body-sized trough.
There was a complex control box at one end, with heavy-duty tubes leading off to
11
another device, in a closet. Temple showed me this as well. lt's an industrial compressor," she said, "the kind they use for filling tires."
"And what does this do?"
The Informant's Perspective: An Anthropologist on Mars
247
"It exerts a firm but comfortable pressure on the body, from the shoulders to
the knees/' Temple said. "Either a steady pressure or a variable one or a pulsating
one, as you wish," she added. "You crawl into it-1 1 11 show you-and turn the compressor on 1 and you have all the controls in your hand, here, right in front of you.11
When I asked her why one should seek to submit oneself to such pressure,
she told me. When she was a little girl, she said, she had longed to be hugged but
had at the same time been terrified of all contact. When she was hugged, especially by a favorite (but vast) aunt, she
felt overwhelmed, overcome
Pl VWOOD CUTTING DIAGRAM: f'v.KE ALL curs ON CENTER
USE TWO SHEETS Of 3/4~ (19mm) AC
USE THIS LAYOlff so
by sensation; she had a sense of
PLYWOOD GRAIN WILL FACE INTHE RIGHT DIRECTION
peacefulness and pleasure, but
also of terror and engulfment. She
started to have daydreams-she
was just five at the time-of a magic
Pl!LlE MJUNT
machine that could squeeze her
-~OARD
powerfully but gently, in a huglike
way, and in a way entirely commanded and controlled by her.
Years later, as an adolescent, she had
seen a picture of a squeeze chute
designed to hold or restrain calves
and realized that that was it: a little
modification to make it suitable for
human use, and it could be her magic
machine. She had considered other
devices-inflatable suits, which
could exert an even pressure all over
the body-but the squeeze chute, in
its simplicity, was quite irresistible.
Being of a practical turn of mind,
Schematic diagram of Temple Grandi n's squeeze machine
she soon made her fantasy come
true. The early models were crude,
with some snags and glitches, but she eventually evolved a totally comfortable,
predictable system, capable of administering a "hug" with whatever parameters
she desired. Her squeeze machine had worked exactly as she hoped, yielding the
very sense of calmness and pleasure she had dreamed of since childhood. She
could not have gone through the stormy days of college without her squeeze
machine, she said. She could not turn to human beings for solace and comfort,
but she could always turn to it. The machine, which she neither exhibited nor concealed but kept openly in her room at college, excited derision and suspicion and
was seen by psychiatrists as a "regression" or "fixation" -something that needed
to be psychoanalyzed and resolved. With her characteristic stubbornness, tenacity,
single-mindedness, and bravery-along with a complete absence of inhibition or
hesitation-Temple ignored all these comments and reactions and determined to
find a scientific validation of her feelings.
11
11
[:
-::',;
248
Chapter 5
Researching People: The Collaborative Listener
Both before and after writing her doctoral thesis, she made a systematic
investigation of the effects of deep pressure in autistic people, college students,
and animals, and recently a paper of hers on this was published in the Journal
of Child and Adolescent Psychopharmacology. Today, her squeeze machine,
variously modified, is receiving extensive clinical trials. She has also become the
world's foremost designer of squeeze chutes for cattle and has published, in the
meat-industry and veterinary literature, many articles on the theory and practice of
humane restraint and gentle holding.
While telling me this, Temple knelt down, then eased herself, facedown and
at full length, into the "V," turned on the compressor (it took a minute for the
master cylinder to fill), and twisted the controls. The sides converged, clasping
her firmly, and then, as she made a small adjustment, relaxed their grip slightly.
lt was the most bizarre thing I had ever seen, and yet, for all its oddness, it was
moving and simple. Certainly there was no doubt of its effect. Temple's voice,
often loud and hard, became softer and gentler as she lay in her machine. 11 1
concentrate on how gently I can do it," she said, and then spoke of the necessity
of"totally giving in to it .... I'm getting real relaxed now," she added quietly. "I
guess others get this through relation with other people."
It is not just pleasure or relaxation that Temple gets from the machine but, she
maintains, a feeling for others. As she lies in her machine, she says, her thoughts
often turn to her mother, her favorite aunt, her teachers. She feels their love for
her, and hers for them. She feels that the machine opens a door into an otherwise
closed emotional world and allows her, almost teaches her, to feel empathy for
others.
After twenty minutes or so, she emerged, visibly calmer, emotionally less rigid
(she says that a cat can easily sense the difference in her at these times), and asked
me if I would care to try the machine.
Indeed, I was curious and scrambled into it, feeling a little foolish and selfconscious-but less so than l might have been, because Temple herself was so
wholly lacking in self-consciousness. She turned the compressor on again and
filled the master cylinder, and I experimented gingerly with the controls. It was
indeed a sweet, calming feeling-one that reminded me of my deep-diving days
long ago, when I felt the pressure of the water on my diving suit as a whole-body
embrace.
As you can tell from this essay on Temple Grandin, the process of asking
and listening collaboratively allows us to gain the perspective of an "other."
Examining our own assumptions and worldviews from the vantage points
of others exposes us to our quirks and shortcomings and cultural biases.
In the process of understanding others, we come to more fully understand
ourselves.
·.•
_·.·
....
_
· ·.
_..•
_
· · ·· r . · . _ ·
..
It'
Gathering Family Stories
;,
~of
Gathering Family Stories
Yet another way to gather dala from an informant is to listen for stories. Temple
Grandin's stories and Teardrop's stories came directly out of answers to their
interviewer's questions. But sometimes stories that are buried and unconscious
offer important information about our lives.
Stories, like artifacts, serve to tell us about our informants' \.Vorldviews and
function as data in our fieldwork. Informa_nts have entire repertoires of stories based on lheir childhoods, their interests, their occupations. Our job as
researchers is to elicit our informants' stories, record them, and carefully analyze what they mean. Researchers who study verbal art think about stories in
these ways:
Stories preserve a culture's values and beliefs.
:y
Stories help individuals endure, transform, or reject cultural values for
themselves.
Stories exist because of the interrelationship between tellers and audiences.
1e
d
d
g
s
l
The most influential kinship stn1cture is, of course, the family. And stories
begin in our families. And to understand someone's culture, we often need to
understand the person's family too. Family stories help us do that. Because
we first hear them when we're young, family stories influence and shape us. In
many cultures, family storytelling sessions are a deliberate way of passing along
values. They are often expected events, almost ritualized performances. Judith
Ortiz Cofer, in a memoir of her Puerto Rican childhood, writes about ho\v the
younger females in her extended family were encouraged to eavesdrop on the
adult storytelling ritual:
At three or four o'clock in the afternoon, the hour of the cafe con /eche, the
women of my family gathered in Mama's living room to speak of important
things and retell family stories meant to be overheard by us young girls,
their daughters .... It was on these rockers that my mother, her sisters,
and my grandmother sat these afternoons of my childhood to tell their
stories, teaching each other, and my cousin and me, what it was like to
be a woman, more specifically, a Puerto Rican woman. They talked about
life on the island, and life in Los Nueva Yores, their way of referring to the
United States from New York City to California: the other place, not home,
all the same. They told real-life stories, though, as I later learned, always
embellishing them with a little or a lot of dramatic detail. And they told
cuentos, the morality and cautionary tales told by the women in our family
for generations; stories that became part of my subconscious as I grew up
in two worlds. (64-65)
249
250
Chapter 5
Researching People: The Collaborative Listener
These stories from Ortiz Cofer's childhood were not merely afternoon entertainment. Her family's stories recorded history and carried instructions about
behaviors, rules, and beliefs. Like the legends, folk tales, and proverbs of specific
cultures, family stories reflect the ways of acting and even of viewing the world
sanctioned or approved by a family. Ortiz Cofer's relatives conserve cultural traditions of their old country, Puerto Rico, and translate them into the "Los Nueva
Yores" culture.
In addition to preserving cultural values, many writers suggest that the act
of storytelling is also an act of individual survival. To endure in our families and
the culture at large, we must explain our lives to ourselves. First we share our
stories, and then we reflect on what they mean. Our own storytelling inemories
teach us about our personal histories. When you think of a family story, try to
decide why it survived, which tellers have different versions, what parts of the
story remain the same no matter who tells it, and how you've refashioned it for
your own purposes.
Family stories are often transformed in oral retellings, hut they clearly
change when they are written down. Although they belong to the oral narrative
tradition, writing them down helps us analyze their meaning and potential relevance to our own lives.
Writing a Family Story
We're not always conscious of how our family stories serve us, nor are our informants.
But they are worth exploring as a rich source of data when we want to better under~
stand our own lives-and the lives of the people with whom we're working. As scholar
Elizabeth Stone, in Black Sheep and Kissing Cousins: How Our Family Stories Shape Us,
writes,
What struck me about my own family stories was first, how much under my skin
they were; second, once my childhood was over, how little deliberate attention
I ever paid to them; and third, how thoroughly invisible they were to anyone
else. Going about my daily life, I certainly never told them aloud and never even
alluded to them .... Those who say that America is a land of rootless nomads who
travel light, uninstructed by memory and family ties, have missed part of the
evidence.
lf[(!!lli!@J~
Recall a family story you've heard many times. It may fall into one of these categories:
fortunes gained and lost, heroes, "black sheep," eccentric or oddball relatives, acts of
terout
ific
•rid
traeva
retribution and revenge, or family feuds. After writing the story, analyze its meaning.
When is this story most often told, and why? What kinds of warnings or messages
does this story convey? For the family? For an outsider? What kind of lesson does the
story teach? How does your story reflect your family's values? How has it changed
or altered through various rete!lings? Which family members would have different
versions?
act
md
our
:ies
I to
the
for
trly
:ive
rel-
our student Teresa Shorter writes, 11 My family on my mother's side came from very poor
roots. My maternal great-grandparents were farmers who struggled by with 13 children.
My paternal great-grandparents raised a dozen children-most were boys who came to
violent ends. Their stories are of struggles and poverty. I think this story is told in my family to teach us of our humble beginnings and to help us be aware of how bad it could be.
And even though we didn't have a lot of money growing up, my cousins and I always had
food to eat, a bed to sleep in, and arms that would hug us-we never had to live in fear.
We always knew we were loved." Teresa's story goes like this:
My Grandmother's Wise Stand
Teresa Shorter
My maternal grandfather was named Daniel Hancock. I never knew him-he died
when my mother was only nine years old. Of his nine children, she was his favoritethis was almost her undoing.
Daniel Hancock was a mean man. He was a moonshine runner in the hills of
southern Virginia. Times were tough and there was never enough to eat. He was a
small, dark man who loved to drink, and when he drank he became a monster. He
would beat my grandmother and the older children, often coming at them with a
knife. They would try to hide when he came home, afraid he would kill one of them
one day. They lived a life of fear and want.
One day a few weeks after the birth of the twins Ray and Fay, my grandfather
was preparing for another run. He had already been drinking. As he often did, he
tried to take my mother riding with him. I think he favored her best because she is
so different from him-most of the other children have his Native American looks,
but my grandmother at nine had bright blue eyes and whitish-golden hair. She took
after my grandmother. Even though my young mother was willing to go "riding,"
my grandmother convinced her inebriated husband not to take my mother on this
run-a run from which he would not return.
Hours later, relatives would be at the door telling my grandmother that Daniel
had been kilted in a car accident. His car had run off the road around the bend a few
miles out. That was the story. However, these were the roads he had run his whole
life-drunk and sober. Truth came out later that he was found with his head bashed
in from behind. We'll never know if it was a robbery or if someone stepped in to save
my grandmother and her children-these were rough times. All l know is that my
mom dodged a bullet that day.
251
250
Chapter 5
Researching People: The Collaborative Listener
These stories from Ortiz Cofer's childhood were not merely afternoon entertainment. Her family's stories recorded history and carried instructions about
behaviors, rules, and beliefs. Like the legends, folk tales, and proverbs of specific
cultures, family stories reflect the ways of acting and even of vie\ving the world
sanctioned or approved by a family. Ortiz Cofer's relatives conserve cultural traditions of their old country, Puerto Rico, and translate them into the "Los Nueva
Yores" culture.
In addition to preserving cultural values, many writers suggest that the act
of storytelling is also an act of individual survival. To endure in our families and
the culture at large, we must explain our lives to ourselves. First we share our
stories, and then \Ve reflect on what they mean. Our own storytelling memories
teach us about our personal histories. When you think of a family story, try to
decide why it survived, which tellers have different versions, what parts of the
story remain the same no matter who tells it, and how you've refashioned it for
your own purposes.
Family stories are often transformed in oral retellings, but they clearly
change when they are written down. Although they belong to the oral narrative
tradition, writing them down helps us analyze their meaning and potential relevance to our own lives.
Writing a Family Story
We're not always conscious of how our family stories serve us, nor are our informants.
But they are worth exploring as a rich source of data when we want to better understand our own lives-and the lives of the people with whom we're working. As scholar
Elizabeth Stone, in Black Sheep and Kissing Cousins: How Our Family Stories Shape Us,
writes,
What struck me about my own family stories was first, how much under my skin
they were; second, once my childhood was over, how little deliberate attention
I ever paid to them; and third, how thoroughly invisible they were to anyone
else. Going about my daily life, I certainly never told them aloud and never even
alluded to them .... Those who say that America is a land of rootless nomads who
travel light1 uninstructed by memory and family ties, have missed part of the
evidence.
ll!tJilli i [of~!Jj
Recall a family story you've heard many times. It may fall into one of these categories:
fortunes gained and lost, heroes, "black sheep," eccentric or oddball relatives, acts of
retribution and revenge, or family feuds. After writing the story, analyze its meaning.
When is this story most often told, and why? What kinds of warnings or messages
does this story convey? For the family? For an outsider? What kind of lesson does the
story teach? How does your story reflect your family's values? How has it changed
or altered through various retellings? Which family members would have different
versions?
Our student Teresa Shorter writes, 11 My family on my mother's side came from very poor
roots. My maternal great-grandparents were farmers who struggled by with 13 children.
My paternal great-grandparents raised a dozen children-most were boys who came to
violent ends. Their stories are of struggles and poverty. I think this story is told in my family to teach us of our humble beginnings and to help us be aware of how bad it could be.
And even though we didn't have a lot of money growing up, my cousins and I always had
food to eat, a bed to sleep in, and arms that would hug us-we never had to live in fear.
We always knew we were loved.11 Teresa's story goes like this:
My Grandmother's Wise Stand
Teresa Shorter
My maternal grandfather was named Daniel Hancock. I never knew him-he died
when my mother was only nine years old. Of his nine children, she was his favoritethis was almost her undoing.
Daniel Hancock was a mean man. He was a moonshine runner in the hills of
southern Virginia. Times were tough and there was never enough to eat. He was a
small, dark man who loved to drink, and when he drank he became a monster. He
would beat my grandmother and the older children, often coming at them with a
knife. They would try to hide when he came home, afraid he would kill one of them
one day. They lived a life of fear and want.
One day a few weeks after the birth of the twins Ray and Fay, my grandfather
was preparing for another run. He had already been drinking. As he often did, he
tried to take my mother riding with him. I think he favored her best because she is
so different from him-most of the other children have his Native American looks,
but my grandmother at nine had bright blue eyes and whitish-golden hair. She took
after my grandmother. Even though my young mother was willing to go "riding,"
my grandmother convinced her inebriated husband not to take my mother on this
run-a run from which he would not return.
Hours later, relatives would be at the door telling my grandmother that Daniel
had been killed in a car accident. His car had run off the road around the bend a few
miles out. That was the story. However, these were the roads he had run his whole
life-drunk and sober. Truth came out later that he was found with his head bashed
in from behind. We'll never know if it was a robbery or if someone stepped in to save
my grandmother and her children-these were rough times. All I know is that my
mom dodged a bullet that day.
251
252
Chapter 5
Researching People: The Collaborative Listener
Family stories like Teresa's preserve family beliefs through morality lessons
with subliminal messages and subconscious instructions. Some family stories
act as cautionary tales, or what writer Judith Ortiz Cofer calls cuentos, to pass
on warnings about behavior to the next generation. Through storytelling, family
tales can be transformed and reshaped to make them fit the teller's needs and
life circumstances. As they transform family stories, tellers can remain loyal to
the family unit but be released from it as well. They can connect us with our
families while allowing us our identities as we reshape them to fit our own lives
and our own audiences.
One Family Story: The Core and Its Variants
When Donna Niday, a professor at Iowa State University, w:is a student in Bonnie's class, she decided to study one of her family's stories. It illustrates a significant idea: that different family members tell one story in very different ways,
and each \Vay reveals something about the teller. Donna was in her forties at the
time, and she was lucky that her mother and sisters were all alive and accessible while she conducted her study. In researching "The Baby on the Roof,"
Donna expected to get different perspectives from the five Riggs sisters-her
own mother and her four aunts who grew up on a farm in the n1ral Midwest of
the 1920s. She began by interviewing her cousins to see what they remembered
about the story, which would have been passed down to them by their mothers. Just as she suspected, each of Donna's cousins told the story with different
details and points of emphasis. She confirmed the aunts' family reputations:
"she was the daring one" or "she was always the chicken."
Donna was realistic enough to know that there would be no "true" version
of the tale, but the story would have a stable core, a basic frame, shared among
the sisters. She also anticipated that there would be many variants, differences
in details according to the tellers. She interviewed each elderly sister in her
home, allowing time to look through family photo albums together and to visit
before recording the stories. As she listened and recorded, Donna gathered both
the core story and its variants.
The oldest sister, Eleanor, who claimed responsibility for the secret family
event, told the core story this way, emphasizing herself and the baby:
I took Mary to the top of the house when Mom went out to work. You
know, all four of us took the ladder-went down to the barn and got the big
ladder. Mom just said to take care of Mary, and I did. I took her everywhere
I went. Mary was six months old. Well, she was born in January. This would
be June, I suppose, when we were doing hay. I knew she'd lay wherever
you put her. And so there was a flat place there on the roof. There wasn't
any danger-I don't think there was ever any danger of her getting away
at all. Yeah, we could see them mowing. If Mom had ever looked toward
the house, she would have had a heart attack to see her kids up on the roof.
One Family Story: The Core and Its Variants
Especially when we were supposed to be looking after the haby. Yeah, well
see, I was ten when Mary was born, so I was ten then, a little past ten. I
should have known better, but it shows that you can't trust ten-year-olds.
I never got punished for that because Mom never did find out.
And so, with Eleanor's version, Donna had the core family story: four sisters spent the day on the roof with their six-month-old hahy sister while their
mother and father mowed the fields. Each sister provided her own variation.
One remembers that the parents were mowing hay; another insists they were
cutting oats. Such details would also change the time of the story from June
to late summer or early fall. The sisters dehate other details. Donna's mother
rejects the idea that they got a ladder from the barn, saying they climbed on a
chair or rain barrel to reach the lower part of the farmhouse roof, which was
accessible from their bedrooms. Another of the sisters tells the story as if she
remained on the ground while the others climbed to the roof. When they challenge her version, this sister admits that she probably did follow the others. She
confesses jealousy of the new baby: until then, she had been the baby of the family. "They weren't worried about me," she recalls. "They were only worried about
Mary, the baby."
Mary's version of the story deviates the most. She claims that she fell off the
roof and that her sisters climbed down and put her "right back on." Because
Mary's variant has no support from the other sisters, she retreats by saying,
"Maybe I just dreamed that, but it seems like I fell off when I was up there. Of
course, I wasn't really old enough to remember."
The Riggs family farmhouse
253
254
Researching People: The Collaborative Listener
Chapter 5
After recording and transcribing all five versions of this story, Donna proposes that the "baby on the roof" story displays defiance of authority and rebellion against rules for these otherwise compliant farm girls. She also thinks
that the story illustrates "pluck and adventure," as no harm was actually done.
Donna admits that her mother and aunts would deny that these stories convey
any meaning other than "pure entertainment" and confesses that any
For information on
analysis of the meaning of these stories is her own. She recognizes
working with family
that each sister embellished the story based on her family reputation,
archives, see Chapter 7,
page 317.
individual temperament, and storytelling ability. Donna's conclusions
are consistent with what we know: that a story has a stable core of
details but also many variants according to the tellers .
'
...................................................................................................................
Discover Core Details in Your Informants' Stories
Use the following questions to uncover the core details of your informants'
stories:
10
What facts are stable?
©
What's the chronology of the story?
~
What characters are key?
What is the central conflict in the story?
What is the theme?
.
Does the story contain a cultural message or lesson?
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
Donna followed several important steps in gathering her family stories,
steps that any fieldworker considers:
@
e
She conducted preliminary research by interviewing the people
involved-in this case, her mother and aunts. Had they all told the same
story, the project would not have illustrated the core and variants inherent
in family stories.
She interviewed her informants in their own home settings, making all
participants comfortable as she recorded and asked questions. She didn't
rush her project; she allowed time for scheduling, visiting, interviewing,
transcribing, sharing the transcripts, and writing her paper.
She triangulated her data in two ways. First, by checking the five stories
against one another, she could see how one story n1ight verify another
or disconfirm it. She shared her work with the sisters as she went along
and afterward so that they might confirm, disconfirm, or add to each
other's stories.
t}
She acknowledged the importance of her informants' participation. In
this case, as they were her relatives, she presented her essay in a family
Gathering Oral Histories
album complete with photos of the old farmhouse as a gift in return for
the time and energy they spent helping her learn how to listen to family
stories. This kind of reciprocity is crucial to the ethics of fieldworking.
s
y
y
s
s
f
Stories that are passed down within a culture help to shape a culture's selfidentity. But it's also true that the variants (each teller's version) of a story can
explain even more. Kno\.ving how the variants differ helps us find clues about
informants, their worldviews, and how a culture has changed over time. If you
are working in a fieldsite, you will want to find any important stories that have
come out of a shared event, an important moment, or a special person whom
everyone knows.
In your own work, it is important to record each teller's version as audio or
video or by careful word-for-word notetaking. Whether you record several versions of one story or several stories about one event, you will need to look for
the unchanging core elements as well as the variants' details. After gathering
stories, analyze your data. Here are some things to look for .
......................................................................................................................
Discover the Variants in Your Informants' Stories
,•
;,
0
Use the following questions to uncover the variants in your informants'
stories:
0
What are the features that change?
©
Why do those features change?
@
What do the variants suggest about the tellers?
®
What do the variants suggest about their audiences?
You may want to find some features that don't fit the other versions, and those
will provide clues about the teller's positions and attitudes toward the story
and the cultural themes that the story contains.
....................................................................................................................
Gathering Oral Histories
An oral history is a life story shared collaboratively with a fieldworker, emphasizing the individual's life against the cultural significance of that life. Cindie
Marshall uses her interviews to support her research on the biker subculture.
But if she were to do an oral history, she would interview Teardrop in depth to
record her whole life's story. Teardrop's role as a biker's companion would be a
major part of her oral history, but the emphasis would be on Teardrop.
In an oral history, the fieldworker gathers real-life stories about the past
experiences of a particular person, family, region, occupation, craft, skill, or
topic. The fieldworker records spoken recollections and personal reflections
from living people about their past lives, creating a history.
255
256
Chapter 5
Researching People: The Collaborative Listener
Anthropological fieldworkers who record an entire life's history as well as
speculate on the relationship between that life and the culture it represents
are called ethnohistorians, and their studies span many years. Over ten years'
time, folklorist Henry Glassie visited, interviewed, and wrote about one foursquare-mile Irish community in Passing the Tinze in Balleymenone. Shirley
Brice Heath, a linguist, researched literacy in the Piedmonts of the southeastern United States for her study Ways with Words, and spent 14 years gathering
data from parents, teachers, and children there. Anthropologist Ruth Behar
traveled back and forth between the United States and Mexico for over five
years and wrote a life history of Esperanza, a Mexican street peddler, in her
book Translated Woman.
However, not all oral histories need lo be full-length ethnographic studies.
There are many national oral history projects, including Save Our Sounds,
Story Corps, and the Veterans' Oral History project, all sponsored by the American Folklife Center of the Library of Congress. A quick Internet search reveals
a range of local, national, and global projects that all use forms of oral history
collection.
Gathering oral history is not new. A successful traditional project in
the United States comes from a group of high school students in 1970s rural
Georgia. There, teenagers documented the stories, folk arts, and crafts of
elderly people in their Appalachian community, writing about moonshine, faithhealing, building log cabins, dressing hogs, and farming practices. With their
teacher Eliot Wigginton, they originally published their fieldwork in the high
school's Foxfire magazine, and later collected these articles into the now classic
Foxfire anthologies (12 books in total), one of which gives us the eyeball story on
p. 244.
In this section, we introduce you to three different varieties of oral histories: one visual, one auditory, and one textual. Our first oral historian, Nancy
Bauserman, is a professor of law and ethics in tbe College of Business at the
University of Iowa and an amateur photographer. Nancy has always been interested in the lives of people who are often invisible and ignored. In one study, she
spoke to workers at University of Iowa Hospitals and Clinics, a large hospital
in Iowa City, recording their perspectives on the value of their work. She also
photographed them on the job. As she writes, her project combined her "love of
taking photos along with a deep appreciation for the myriad kindness people do
for others every day." Nancy's collection, called "Taking Care: A Recognition of
Good People Doing Things," is a permanent exhibit on the walls of the hospital
and has also been published as a book. Its power, we believe, comes from the
artful combination of photographs of ordinary people in their workplace supported by the words of each informant. Their words are ones you might not
always have the chance to hear.
We present here two of her seventeen oral history portraits, this one of Allen
Reed, a custodian, followed by one of Mona Ibrahim, a dietetic clerk, both of
whom talk about how they feel about their jobs.
Gathering Oral Histories
swell as
presents
~n years'
ne fourShirley
utheasttthering
1 Behar
1ver five
~ in her
studies.
:;ounds,
Amerireveals
history
iject in
Is
' " ' ' ,,,,, ,.,
,~,
"" ""
,~,
'"'" "'"' """ ""''
>~"""
0000'' '''" ,.,.
,.~""''~'""-''
.o' ,, "·'""'
"
257
~··"·~~'""""
~'.
Taking Care
Nancy Hauserman
Allen Reed, Custodian
work second shift with the project crew. I pick up already bagged trash and make
1
it disappear. I strip and wax hardwood floors 1 clean carpet, move furniture before
and after meetings, burnish and wax floors. When I'm House Man 1 I do anything
for anyone as needed.
Any requests I get from patients and visitors, if I can take care of it, it's with
a positive attitude like "I'm gonna take care of this for ya." In fact I see somebody
that 1s, well, you just know these people are lost. I will walk up and say, 11 Excuse me
Fm Allen Reed. Can I give you some assistance? 11 They may say, "We're looking
for Emergency." I never give directions to Emergency; I always walk them to the
Emergency Room, because if you're asking for the Emergency Room, you have
other things on your mind. And when some folks say, "Well, you're not going to get
in trouble are you?" I tell them, "It's not a problem, anytime you folks need help
rural
·afts of
" faithh their
te high
classic
:oryon
f
I*
histoNancy
at the
1 inter:ly, she
)Sp ital
le also
ove of
ple do
ion of
>Sp ital
m the
u
i:
#:R-,
~;
~,:11!
ii_
¥:
t:'
~i)if
~0
i;:'.
•E
I•
~sup-
1t
'2
not
>c
u
z•
Allen
'th of
§"
""~
Allen Reed, custodian, University of Iowa Hospitals and Clinics
[t
~t
i:
;;"'
:,
li;·:
)';
};
'~'l'·
i-:<
258
Chapter 5
Researching People: The Collaborative Listener
that can be my top priority beside what I'm supposed to be doing." So I never take
no for an answer; I never say I can't do it, I always look into it, and I'll work on a
problem as long as it takes to get it done.
I'm here every day because I enjoy helping people. There's no time that I
haven't been able to give directions or take someone where they needed to be or
help a patient or a visitor out. It makes you feel like you've accomplished some-
thing at the end of the night, especially when I'm House Man and I get a lot of different calls during the night from patients or visitors. It makes a difference.
I always try to say, is this what I would want?
Mona Ibrahim, Dietetic Clerk
I take patient orders on the phone and check to make sure they are ordering
according to their diets. If the food isn't on their diet, I try to suggest what they
could have instead. I call patients if they don't call in for meals for a while. We use
computers now, but for years it was all done by paper.
I think what I do makes it easier sometimes [for patients and families]. When
patients are ordering, sometimes the family is not there. They call and ask how
Mona Ibrahim, dietetic clerk
r
Gathering Oral Histories
1ever take
·kon a
hat I
Jto be or
somelot of dif:e.
1
,ring
1t they
~-We use
;],When
khow
they should order,,, they don't know their diet sometimes so we try to tell them
what they can have and what they can't. At the same time, sometimes they just like
to talk to me on the phone. When speaking to someone for a while and they call
you,,. the way they answer, you can tell that they feel there is someone there that
they know.
You think about it as if you were in their place. They can be comforted by
someone to speak to.
, "" o•
oo•••o~••••••••••••••
••••• ••••••••••••••• •••• ,, , ••••, •••••
00000000000000 • • • • • • • • • • • • • • • •
Recorded-rather than written-oral histories have their own strengths. In
them, you can experience a range of linguistic features: people's actual voices,
their accents, hesitations, intonations, laughter, and the overall rhythm of their
speech. How many times do you wish you could bring back the voice of someone
you've lost? An audio interview offers us the ability to do that. As the researcher,
you choose among the recorded data about how much or how little of the informant's language you want to include. You also choose what form-electronic
or print-you will present it in. No matter what technology you use or whether
your recorded oral history ends up as an essay or an mp3 file, you will need to
decide how to incorporate your informants' words with your own. Some oral
histories present only the informants' words. But note that the editor of those
words is usually the researcher. As with other data collections, most oral historians record far more words than they use, whether the final presentation is on
paper or in an audio file. It may seem easy to simply record someone else's life
story, but the responsibility for arranging, organizing, and editing it is still yours.
One of our favorite examples of auditory oral history is the StoryCorps
project. It began in 2003, became associated with the American Folklife Center
at the Library of Congress, and is now a regularly featured series on National
Public Radio. Dave Isay, the founder of StoryCorps, began with a booth in
Grand Central Station in New York. On the day it launched, Studs Terkel, a great
oral historian of the twentieth century, proclaimed, "Today we shall begin celebrating the lives of the uncelebrated." Since then, the project has added many
permanent and mobile recording studio-booths around the country. You could
go to one with your boss or favorite teacher, for example, and spend 45 minutes talking with the assistance of a facilitator. You'd end up with a broadcastquality CD for yourself-as well as contributing your story to the archive at the
American Folklife Center. It is from these CDs that StoryCorps has edited threeminute oral histories and organized them into a range of topics. You can sample
them at http://storycorps.org/listen.
We'd like to share one example from Isay's first collection, Listening Is an
Act of Love: A Celebration of American Life from the StoryCorps Project (2007),
recorded in Portland, Maine, between Joyce Butler, 73, and her daughter, Stephanie Butler, 47. In it, Joyce talks about her mother, who was a welder and single
mom during World War IL
259
--...
,~..
I'
I
~
E
[,
k
~
-------------~~-
260
Chapter 5
Researching People: The Collaborative Listener
,., ......... ,,.~,."'''""'"'""'0'"''"'"''"''''''''""""~"'""'"''''''""'"'"''''"''''''''00•~······
Listening Is an Act of Love
Joyce Butler, 73, interviewed by her daughter, Stephanie Butler, 47
Joyce Butler: My parents divorced in 1941. In those days divorce was not as common as it is now. A divorced woman was called a "grass widow," and grass widows
were scorned. There was a stigma attached, which is why it took such courage
for my mother to ask for a divorce from my father. We moved to Maine Avenue in
Portland.
Even though my father paid my mother money every month, it wasn't enough.
She had to go to work, and she had not finished high school. So she worked in the
laundry, and she finally got a job at Montgomery Ward department store on Congress Street. And we kids were more or less on our own. That was not a happy time
for me-I missed my mother.
During the war, the shipyard had begun to function in South Portland, and
these young women would come into the store, all dressed in these big boots and
rough overalls. And they would have checks of six hundred dollars to cash for their
shopping. She finally asked, "Where do you work that you make so much money?"
They said, "In the shipyard." So my mother went over and tried to get a job. The
man who interviewed asked if she wanted to be a welder or a burner, and my
11
mother said, "Which pays the most?" And he said, "A welder. And she said, "Thafs
what I want to do." And he said, "Oh, a mercenary, huh?" And she said, "No. I have
four children to take care of.
11
Her shift was midnight to 6:00 a.m., so she could be
home with us during the day. I remember her dressing in
that heavy clothing and big boots- men's clothing. Once
she fell and hurt her ankle, and they brought her home
in the middle of the night, and she was weeping. It was
awful.
It was bitter cold in the winter, going into the bowels
of those steel ships. They had to wiggle into narrow crawl
spaces and lay on their backs and weld overhead. She was
very thin in those years, but I remember her neck and her
chest, all spotted with burn marks from the sparks. They
had to wear special goggles, but even so, sometimes they
would have a flash condition in their eyes. She suffered
from that, and they had to take her to the hospital once.
After the shipyard closed, she went back to Montgomery Ward and worked all
day. And at five o'clock when she got out of Montgomery Ward, she got on a bus
and worked in the S. D. Warren Paper Mill from six o'clock until midnight. Came
home, got up in the morning, and went back to Montgomery Ward.
Stephanie Butler: Bless her heart.
Joyce: My mother wanted to keep us together as a family. She was determined.
Gathering Oral Histories
l.
e
18
j
dr
"'
261
Our final example is of a textually-rendered oral history and comes from
the Great Depression. In the United States of the 1930s, writers were among
the many unemployed. The government sponsored the Federal Writers' Project,
which put writers to work as interviewers. Among them were Claude McKay,
Richard Wright, Saul Bellow, Loren Eisley, and Ralph Ellison. This project's goal
was to record the life histories of ordinary American people whose stories had
never been told: carpenters, cigar makers, dairy people, seamstresses, peddlers,
railroad men, textile workers, salesladies, and chicken farmers were among the
informants.
One collection of these life histories, These Are Our Lives (1939), assembled
by R. R. Humphries for the Federal Writers' Project, includes the life story of
Lee Lincoln, a man who learns to read and write as an adult. The ........................................ ..
fieldworker, Jennette Edwards, inserts her own observations and Read Jennette Edwards's
description into the interview while quoting Lee's words directly as piece, / Can Read
and I Can Write, at
she collected them. You can access the excerpt on our Web site. As you bedfordstmartins.com/
read this piece, remember it was written in 1939, when black Ameri- fieldworking, under
cans were called Negroes, when jobs were difficult to get, and $65 a Professional Essays.
month was a decent living. You may want to think about other cultural assumptions, attitudes, and beliefs that have changed in the past decades.
s
Box24
1
e
s
vi
as
y
ill
Starting an Oral History
Many oral histories today are gathered from ordinary people who have lived through
extraordinary times and experiences. Contemporary informants can share their life
histories and experiences from, for example, the women's movement, the civil rights
movement, the Holocaust in Europe, or the end of apartheid in South Africa. Other oral
histories can record the everyday life during an occupation that no longer exists. Studs
Terkel's collections of occupational stories, Working, and of Depression stories, Jn Hard
Times, offer examples of short histories from real people whose voices are seldom heard
or recorded.
Possible projects for an oral history are limited only by your imagination and access
to the people you wish to interview. Many people begin an oral history by interviewing
their relatives, friends, or teachers about living through a particular era or a time of personal struggle that resulted in dramatic life changes. Perhaps you know someone who's
lived through a major catastrophe (such as an earthquake or a flood) or someone who's
been caught personally in a political or social entanglement (like war, bankruptcy, or
discrimination). Such people make good subjects for oral history.
•
Box24
continued
Many local and compelling oral history projects can emerge in unexpected places
and on unexpected topics. Think about someone you know who has a particular skill,
such as cooking ethnic food, whittling wooden figures, weaving, or embroidering, or an
unusual hobby such as clogging, playing the bagpipes, or raising llamas. Someone's lifelong passion can yield fascinating oral histories. Good subjects for oral histories also hide
in places where people spend their time alone-for example, in garages tinkering with
engines or in antique shops restoring items others have discarded.
If you're interested in pursuing an oral history project, spend some time interviewing
someone who fascinates you. As with any other interviewing project, all of the informa~
tion you record in your field notes may not be what you 1 1\ use in a final writewup, but you'll
need to record it nonetheless. Your choices of details will help you feature your subject
without distracting the reader from the informant's own words. An elaborate oral history
takes a great deal of time. We hope that what you'll learn by reading and writing oral history is that it's important to foreground the informant and her words and to background
the topic, yourself, and her surroundings.
Laura Carroll conducted a series of oral history interviews with Mexican women who
had emigrated to the southeastern United States. Laura volunteered to teach English as
a second language in a public library program, where she met one informant, Luis, who
led her to the immigrant women who offered their oral histories. Because of her interest
in women's issues, she felt drawn to interviewing women in their new American homes.
This excerpt is from her first visit with Estela NUf'lez.
Estela Nunez, La Primera Visita/The First Visit
Laura Carroll
I called Estela, and she invited me to come to her house. After much confusion over
her directions, I finally just wrote down her address and looked it up on MapQuest.
I brought my friend and consultant, Luis, to our first meeting so he could interpret
and also because he's fun to have around in a new social situation. When we arrived
at Estela's house, Luis noticed the new truck in the driveway and commented that
her husband must be getting paid \.vell. The house was small, new, and suburban,
spotless, with newwlooking furniture. A 11 telenovela" -a Spanishwlanguage soap
opera-was on the television.
Estela invited us in and introduced us to her three boys, who seemed excited to
have guests. The youngest boy, who looked about five, opened his eyes wide when
he saw me. In English, he shouted, "You're GIANT!" I laughed.
Estela invited us to the table to eat her tacos de polio. She had made a pot of
beans and some spicy red salsa to go with it. The smell and taste transported me to
262
Mexico. I had to stop eating for a second to let the dense meal settle. "Come, come,"
Estela's melodic voice exclaimed. Then I remembered how it's done in Mexico. You
eat until lt's gone.
After listening to Spanish flow at light speed between Estela and Luis, I began to
record her story.
Here are a few excerpts from several of Laura's interviews with Estela that eventually
became her oral history. Estela says:
"I grew up in the small town of El Rosario in Guanajuato. I went to La Primaria but
left in the sixth grade to work at home. Girls and boys often leave school at this age
because the family needs their help at home and in the fields. My father died in an
accident when my mother was 28 years old, so we were needed at home more than
usual. My mother never remarried because she was afraid that a new husband might
treat the children badly....
"Jn the morning, we got up and did las labores de la casa. We washed the clothes
by hand and brought in milk from the cows .... All of the women work inside
the house. Some work outside of the house, but very few. It is hard for us to get
accustomed to the work life here; it is a big change for us ....
"I'm always working, always taking care of the children, always doing housework. I'm muy activa. I like working outside the house. I clean houses three days
a week. I don't need to speak English; I already know how to clean. Paco a poco
I am adjusting to life here. I like it here. I taught myself to drive and went to the
DMV without my husband's permission and got my driver's license. I am learning
English ...."
Estela is also learning how to cook vegetarian. She asked me to go to the store
with her to help her find a type of oil she had heard about that is supposed to
be healthier than corn oil. I expected her to take me to a supermarket, and I was
surprised when we arrived at the local cooperative health-food store where I
used to work. She took me by the hand and led me around the store asking me
what all of the labels on the food meant. "What is organic?" "What is tofu? "I've
heard of soymilk. What is it?" 111 want to be flaca like you! 11 When I answered
each question, she got excited and said, 11Aaaaaah 111 and put the item in her
basket.
During our last visit, Estela told me that she'd be going to El Rosario with
her three children this sumn,ier for two months. Her husband would stay here
and work, although he's very sad not to be going. I asked her how she's getting
there, and she said she'll fly, but she'll have to come back another way because
she doesn't have valid immigration papers. She said that she'll find friends to
help her. She seemed concerned about the possibility of not being allowed back
in the United States, but I could tell that the need to see her mother, whom she
had not seen in five years, far outweighed any problems she might have at the
border.
When I asked Estela if she likes living in the United States, she replied, "Ah, sf, es
11
muy bonito, pero extrafio mi tierra. Extrafio mi Rosario."
263
264
Chapter 5
Researching People: The Collaborative Listener
FieldWriting: Using Character, Setting,
and Theme to Create a Portrait
To bring their informants to the page from a pile of data, fieldwriters must pay
close attention to informants' personal characteristics and surroundings and
write about details that relate to the overall themes they want to highlight.
Creating verbal portraits means studying your fieldnotcs, selecting your most
relevant details, and drafting sentences that portray each informant against a
cultural backdrop. In this chapter, the ficldwriters who give us portraits write
them on the basis of carefully gathered observations from their fieldnotes and
other sources and their own interpretations confirmed by their data.
Details of Character
Choosing the details to describe an interview with an informant is hard work.
Fieldwriters must gather and record far more information than they will ever
use because during data collection they don't know the themes they'll eventually
want to highlight. Written portraits of an informant require noting the same
kinds of character details that fiction writers use: physical features, material
artifacts, body language, oral language patterns, and personal history. But those
details are borne in fieldnotes, interviews, artifacts, and documents.
Physical features (vine-shaped tat loo around wrist; brown hair with gray,
neatly pulled back)
&
Material artifacts (new Lee jeans, pale yellow sweater)
Body language (sitting alone drinking beer, shooting pool)
Oral language patterns (Alice and Christie discuss the bikers' treatment of
women)
" Personal history (goes to Ralph's every day, moved from Michigan three
years ago)
Details of Setting
r.
When fieldwriters paint verbal portraits, they also create a backdrop for their
informants. In Chapter 4, we discuss how writers present landscapes or what
we call "verbal snapshots." Setting details must be organized from notes about
time, place, weather, color, and other sensory impressions at the fieldsite.
To bring her reader into Ralph's, Cindie Marshall moves from exterior to
interior. She selects details of texture and space to represent Ralph's Sports Bar
outside, focusing first on tl1c parking lot.
As she moves inside, she describes tastes and smells in an atmosphere of
smoke, stale beer, cigarette ashes, and body odor. Cindie also listens to the
sounds from the jukebox. Her selection of details to present an image of Ralph's,
a so-called sports bar, defies her earlier assumptions about such places.
FieldWriting: Using Character, Setting, and Theme to Create a Portrait
Details of Theme
pay
and
ight.
nost
1st a
Nrite
, and
.vork.
lever
tu ally
same
Lterial
those
it of
~ee
or their
Jr what
~s about
:erior to
orts Bar
phere of
.s to the
'Ralph's,
Fieldwriters must choose details to support the themes they want to highlight.
In fieldworking, themes don't emerge directly from lists in fieldnotes, words in
transcripts, or library books and collected artifacts, but such sources suggest
them. Themes do conze from active interpretation of your data, as you study
it, triangulate it, organize it, reflect on it, and write about it. Themes arc bigger
than the actual details you record, but those details, as they cluster into categories of data and images from your observations, generate larger interpretations.
For example, in her study of Ralph's Sports Bai; Cindie's themes work off
of the contrasts she observed within the subculture in the biker bar. She arrives
at Ralph's \Vith a mental image of what a sports bar is like, and in1111ediately
that image contrasts with the reality of this biker sports bar. As she spends
more time and takes more notes, Cindie sees other contrasting themes within
this subculture .
Another set of contrasting themes is that of bikers as a community of independent people who "do their own thing," but Cindie sees that within this subculture, they come together only to be separated: "I also wanted to know why
three groups in a bar would come together in a place just to be segregated." Cindie interprets still another contrast when she notices that the biking subculture
includes many women. Her fieldnotes report 13 people at the bar, 8 men and
5 women. But this inclusion is deceptive when her informant Teardrop describes
her life as a biker \voman: "What Teardrop had described \Vas sheer abuse, and
she wore that abuse both on her face, in the shape of a teardrop, and in her
smile, which \Vas darkened by missing teeth."
Field\.vriting is a skill that requires close observation, careful documentation, and the rendering of data into thick descriptions of informants within
their cultural spaces. To be an accurate and sensitive fieldwriter, you'll need to
manipulate your multiple data sources, call on your informants' voices, examine
your reflective writing, and craft a text so that it will give your reader a sense
of participating in the fieldwork you've experienced. As we collect data about
people, we must continually look over what we've gathered in ter1ns of ourselves, our informants, and the information's ineaning against the larger backdrop of our research. In addition to carefully retelling our inforn1ants' stories,
we need to ensure that we present the narrative situation fairly, including our
own roles in it. In his book Writing the New Ethnography, H. L. "Bud" Goodall,
\vhose work you read in Chapter 2, offers some helpful questions for researchers to ask themselves as they write up interviews, oral histories, and storytelling
situations (106-7):
1.
What is the context, and where arc you in this scene? What is the nature of
the relationship between you and your informant?
2.
What's the meaning behind the recorded vvords? What influences your
informants' and your own fixed positions? Subjective positions? What
power relationship exists bet\veen you and your informant?
265
266
Chapter 5
3.
Researching People: The Collaborative Listener
What do you hear in the way the informant speaks? What are the gaps? The
tones? The rhythms? The emphases (of both you and your informant)? (For
more on this idea, see Chapter 6, "Researching Language: The Cultural
Translator.")
In the process of collecting data about people, we must continually look
over what we've gathered in terms of ourselves, our informants, and its meaning against the backdrop of our research. We bring to our narratives about our
informants as much as they bring to the data we've collected about them. In
addition to rendering careful verbal portraits, we need to break down the conversational context to ensure that we present it fairly.
Box25
Writing a Verbal Portrait
JENNIFER S. COOK. RHODE ISLAND COLLEGE
Just as a verbal snapshot (Box 17 in Chapter 4) asks you to synthesize data from your fieldnotes, a verbal portrait asks you to synthesize data from your interviews. It is another way
!i,'.
11
I::
I';
11
II
i!
I'!!'.
to make sense of your data.
A verbal portrait asks that you use data from your interview to paint a portrait of your
participant in a brief piece of writing (500 words) that is rich with description, illustrative
quotes (carefully selected), and thoughtful (but spare) commentary.
In crafting your portrait, you are making choices about what to include and exclude,
about what to reveal and what to ignore. These kinds of editorial choices are a way for
you to interpret your interview data. Additionally, you need to include some evaluation or
assessment (analysis) of the person and his or her role in the subculture you are studying.
In asking yourself, "What was the essence of this interview?" you also need to ask, "What
is the essence of this person?" Both of these questions are central to the kinds of analysis
you should include in your verbal portrait.
To craft your verbal portrait, you need not only interview data and analysis (an interview
transcript you have analyzed) but also a description of your participant. Your job here
is to create an image in the minds of your readers. It also helps to have "supporting
data" -that is, details about the time of day, the weather, the location of the interview, and
The
~For
a description of the space, as well as other significant details about context. Once you
have these essential ingredients, you can begin to sketch a portrait of your participant.
Remember that your overarching goal here is to capture the essence of the person, his
words, and what he means to the fieldstudy.
look
.eant our
n. In
con-
Once you have sketched your portrait, share it with a colleague or classmate to see if
you have successfully communicated the essence of this person and his role in the subculture. You may ask your reader to keep these questions in mind as she reads.
0
@
©
w
5
r
nd
What images or details from this piece create the most vivid impressions?
What "clues" do you get from this piece about the essence of this subculture?
What clutter can I omit to make this piece stronger or clearer?
What would you like to know more about? Where can I provide more detail and
information?
Gabby Lopez's fieldwork project, "Head in the Clouds," is a study of young, aspiring fine
artists in and around Providence, Rhode Island. As a fine arts major, Gabby wanted to
explore the complexities of being a fine artist in a society that, from her perspective, does
not value artists or their contributions to the common good. During the course of her
fieldwork, Gabby discovered that, though fine artists struggle financially, they find happiness in each other's company and safety in the "art world." This is Gabby's verbal portrait
of one of her key participants; it is also the source of the title of the fieldstudy.
Brendan Kennedy sat deep in a chair that more resembled a red, cushioned bowl. His
long, thin legs dangled off the front end of the chair; his elbows lay on the side of the
chair and his hands rested, laced together, on top of his stomach. He had no problem
getting comfortable despite the difficulty of lowering himself into the unstable bowl.
His brown, short hair topped his thin, pale frame. He sat patiently, looking around
the large room, focusing on the nearby chalkboard wall and the elaborate chalk
drawings and vulgarities scribbled on it.
I met Brendan in high school. He was popularly known as the hyperactive boy
who openly expressed himself a bit too much. In his later high school years, Brendan started to express himself through art. He joined a few art classes and became
involved with Riverzedge Arts Project, where he and l met. He is now a first-year student atthe Art Institute of New England; he is 18 years old. More experienced artists
might consider him to be fresh meat.
I asked Brendan to visit me so I could ask him a few questions about his aspiration
to be a fine artist. Sitting in my red chair, Brendan patiently answered my questions.
He seemed to be waiting for things to get interesting; I was worried that I was boring
him. Soon enough, Brendan got excited about something: "I'm affected by things
that other people aren't," he said. "I understand things differently. I know color, and I
can easily see how something is composed when a non-artist probably can't." I asked
Brendan how a non-artist might see him and other artists like him. He took a second
to think, then confidently, proudly, he said, "Like our heads are in the clouds."
•
267
Box25
continued
Brendan is not the only one to experience this "difference." The very nature of
making art is the opposite of the nature of American culture and society today. I
am not surprised that artists like Brendan feel a little different, a little separate from
society. Artists value slowing down and looking closely, while society teaches us to
generalize and judge as quickly as possible. Artists value work that is creative and
visionary, while society values work that pays a high salary. Artists value representa~
tion and interpretation, while society values the "bottom line." These thoughts drove
me to ask Brendan Kennedy his views about feeling different. He replied, succinctly
and assuredly, 11 !1 m kind of okay with it ... art isn't about making money."
•••••• 0 •••••••••••••• 0 0
""~
••••••••••• 0., 0' •••
"~~"
•••• '
• • 0 • • • • • • 0 • • • • • • ,, 0 • • • • • • •
~"
••••••
THE RESEARCH PORTFOLIO:
Reflective Documentation
••• 000000000
0000~00000000
00000
0000000•~~••; ;oo" o oo
00000 0000'
ooo
000000000 •>0000000000000
Whether you've chosen to learn to research people by continuing your major
fieldwork project or by trying separate short studies, we hope you've seen that
the researcher-informant relationship is a symbiotic one. It is full of interaction,
collaboration, and mutual teaching and learning. In this chapter, we focus on
how people are a dominant source of data in the field-through family stories,
remembered histories, responses to questions, the personal artifacts they consider important, and your own observation of them.
The research portfolio is an important place to record, keep track of, and
make sense of the skills you've learned, the routines and organization you've
used, and your responses to readings that illustrate how other fieldworkers have
written about their work. Listing and categorizing your research processes illustrate your progress as a fieldworker. We like to call this charting process "reflective documentation."
For your portfolio, try listing, mapping, outlining, or charting the skills
you've learned and the variety and amounts of material you've gathered. You
may want to document each of your projects from this chapter separately: family stories, oral histories, and interviews. Or if you are working on a large project, use the questions to document what is relevant to your project. Here is an
outline framed around fieldworking strategies. You might choose a few questions under each category, try to work \Vith them all, or document your work in
a different way:
I.
Prepare for the field.
A. What did you read to prepare for your fieldwork? Where did you find it?
B. What did you learn from reading the fieldwork of others?
268
The Research Portfolio: Reflective Documentation
1ed
C. How did you select your informants? How did you prepare before
meeting them?
D. How did you gain access to your informants and your fieldsites? Were
there any problems? What might you do differently next time?
E. What assumptions did you have going into the field? How did you
record them? What did you expect to see, and what did you actually
find?
IL
Use the researcher's tools.
A.
How did you record your fieldnotes? Did you separate observational
notes from personal notes? Did you invent your own method for organizing your fieldnotes?
B.
What equipment did you use? What would you want at hand if you
could do this work again?
C.
How did you transcribe your recordings?
D. What interviewing skills did you develop? What skills would you like
to work on?
E. What different types of data did you gather? Print sources? Cultural
najor
1 that
::ti on,
us on
ories,
, con-
artifacts? Stories and interviews?
III.
A.
Which initial impressions turned out to be part of your final piece?
Which ideas did you discard?
B. What strategies did you develop to categorize your data? Did you use
patterns that were linear? Thematic? Chronological? Abstract to concrete? Concrete to abstract?
:, and
rou've
;have
, illusceflec-
Interpret your fieldwork.
C. What strategies did you develop to analyze your data? What didn't fit?
D. What is your favorite piece of data - or data source - and why?
IV.
Present your findings.
A. What decisions did you make about writing up your fieldwork?
skills
L You
B. How much of your voice is in the final project? How much of your
informants' voices?
: fam-
C. What details did you select to illustrate key points to bring your informants to life?
' proj~ is an
ques·ork in
ndit?
D. Did you use subheadings to guide your reader in your final paper or
some other way to organize your material?
You might decide to use these questions to help you write an essay or commentary about your process as a researcher.
Wherever people interact in the same space, you have an opportunity to
look carefully at artifacts, personal histories, traditional stories, and ways of
behaving together and alone. But simply describing people as you see them does
269
270
Chapter 5
Researching People: The Collaborative Listener
not produce sufficient data for a fieldstudy. You must use many different ways
to gather their perspectives on their experiences- and do so in their voices. As
you gather information about others, you'll also need to record your own feelings, responses, and reactions as you learn about them and your understanding
deepens and shifts. Eventually, your responsibility is to the informants' voices,
perspectives, stories, and histories-as much as it is to your own.
rooA1
Reflect on Researching People
lr!i1sW\
Try listing the people who are present at your fieldsite. Use the following
questions to guide your exploration of how they interact:
0. 0 0. 0 ••• , . " •••
v.
0-0 • • • ,
~~
•• " ••• " 0
~
••• ;
• • ,, • • 0 " " • • • 0
~··
0 •••• " 0 0.,". 0
~·
•••• 0 ••
""~·
••
FIRST How many people are there? Have you talked to each of them? What
connections and disconnections exist between them? How do they help each
other? How do they define their roles within that culture-for themselves and
for others? Who defines the roles for them?
• • " . " 0 0 • • • • • 0 . > • • • • • • , . 0 0 • • ' " " 0 • • • • • 0" ' " '
•<> ••
0 •• 0 0. 0." 0 ••••
·~
•• " 0 ••••• 0 ••• 0 ••• 0."
What artifacts or stories have more than one person discussed
with you? How have those artifacts or stories helped explain the culture or
those people's places in it? How did each person talk about the artifact or tell
the story? How was each story different? What was the same about the stories?
SECOND
,, ""." >." ••• ". ".
0.' 0
·~"
••••• ,
• • • • "" •• " 0 •••• " 0 ••• "
·~
0 •• " •••• " 0' ••••••• ". 0
~
• • 0 " • • • 00
THIRD Whose oral histories would you like to gather? If you had time for a
much longer study about your fieldsite, who else would you interview? What
stories or artifacts would you want to understand? Whose personal histories
would probably yield interesting information for you?
Whether or not this is your first fieldstudy, the skills of asking and listeningof collaborating in conversation with others-are lifetime skills for personal
interactions with family, with friends, with coworkers1 and in new cultural
contexts. Practicing these skills, reflecting on what you're learning, and writing
from other people's perspectives will strengthen your ability to talk, analyze,
evaluate1 and become a more active and knowledgeable citizen.
ar
in
al
gt
ev
or
le:
W•
vi:
ex
n<
al
to
br
fit
Cl
la
ot
Download